Search Results

Search found 18064 results on 723 pages for 'custom panel'.

Page 665/723 | < Previous Page | 661 662 663 664 665 666 667 668 669 670 671 672  | Next Page >

  • xterm not wrapping text properly

    - by mulllhausen
    I'm configuring both my gnome-terminal and xterm columns (i still haven't picked which of these I will be using) and I have a couple of issues I would like to fix: the typing area seems to be smaller (fewer columns) than the display area the typed text is not wrapping to the next line when it reaches the end - it just continues back around on the same line, overwriting the prompt (i have set a custom bash prompt with PS1 in case this is relevant) $ lsb_release -a No LSB modules are available. Distributor ID: Debian Description: Debian GNU/Linux 7.1 (wheezy) Release: 7.1 Codename: wheezy $ echo $TERM xterm $ stty -a [peter@pc ~] $ stty -a speed 38400 baud; rows 52; columns 126; line = 0; intr = ^C; quit = ^\; erase = ^?; kill = ^U; eof = ^D; eol = M-^?; eol2 = M-^?; swtch = M-^?; start = ^Q; stop = ^S; susp = ^Z; rprnt = ^R; werase = ^W; lnext = ^V; flush = ^O; min = 1; time = 0; -parenb -parodd cs8 hupcl -cstopb cread -clocal -crtscts -ignbrk brkint -ignpar -parmrk -inpck -istrip -inlcr -igncr icrnl ixon -ixoff -iuclc ixany imaxbel iutf8 opost -olcuc -ocrnl onlcr -onocr -onlret -ofill -ofdel nl0 cr0 tab0 bs0 vt0 ff0 isig icanon iexten echo echoe echok -echonl -noflsh -xcase -tostop -echoprt echoctl echoke $[peter@mine ~] $ # the column width only goes up to here ------------------------------------------------> the results are identical in both the xterm and in gnome-terminal 3.4.1.1 and as you can see, the output of the stty -a command goes right up to the edge of the screen, while the typing does not go that far. I have found that I can get the desired result by setting the columns to a very large number, eg: $ stty cols 1800 this fixes both problems. Is it the right way to go about solving this problem? Will this "break" any of the output from programs? So far I have tried top and stty -a and these seem OK. more info as requested in the comments i found that if i cat some input into a file then the columns actually strech the full width of the terminal window: [peter@mine applications] $ cat > /tmp/asd aaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaasssssssssssssssssssssssssssssssssssssssssssssssssssssssssssqqqqqqqqqqqqqqqq qqqq does this imply that it is actually bash that is restricting the number of columns and not the terminal? if so then how to alter the number of columns in bash?

    Read the article

  • How to publish an ASP.NET MVC application to a free host

    - by Lirik
    Hi, I'm using a free web host (0000free) which supports ASP.NET MVC, but it uses Mono. This is the first time I deploy an MVC application, so I'm a little confused as to where I need to deploy it. I have Visual Studio 2010 and I used its Publish Feature (i.e. right click on the project name and click publish) and I tried several things: Publish method: FTP to the root folder. Publish method: FTP to the publich_html folder. Publish method: File System to the root folder. Publish method: File System to the publich_html folder. Publish method: File System to a local directory on my computer and then FTP to root and also tried the public_html folder. I went into the cPanel (control panel) to try and see if ASP.NET has to be added/enabled for my web site, but I didn't see anything there. I can't browse to Index.aspx nor can I redirect to it from index.html (as suggested from other posts on the host forum), right now I have a link from index.html to Index.aspx but it's not working either (see http://www.mydevarmy.com) I've also tried renaming Index.aspx to Default.aspx, but that doesn't work either. The search utility of the forum of the host is somewhat weak, so I use google to search their forum: http://www.google.com/search?q=publish+asp.net+site%3A0000free.com%2Fforum%2F&ie=utf-8&oe=utf-8&aq=t&rls=org.mozilla:en-US:official&client=firefox-a I've been reading Pro ASP.NET MVC Framework and they have a chapter about publishing, but it doesn't provide any specific information with respect to the location of publishing, this is all they say (and it's not very helpful in my case): Where Should I Put My Application? You can deploy your application to any folder on the server. When IIS first installs, it automatically creates a folder for a web site called Default Web Site at c:\Inetpub\wwwroot\, but you shouldn’t feel any obligation to put your application files there. It’s very common to host applications on a different physical drive from the operating system (e.g., in e:\websites\ example.com). It’s entirely up to you, and may be influenced by concerns such as how you plan to back up the server. Here is the exception I get when I try to view my Index.aspx page: Unrecognized attribute 'targetFramework'. (/home/devarmy/public_html/Web.config line 1) Description: HTTP 500. Error processing request. Stack Trace: System.Configuration.ConfigurationErrorsException: Unrecognized attribute 'targetFramework'. (/home/devarmy/public_html/Web.config line 1) at System.Configuration.ConfigurationElement.DeserializeElement (System.Xml.XmlReader reader, Boolean serializeCollectionKey) [0x00000] in <filename unknown>:0 at System.Configuration.ConfigurationSection.DoDeserializeSection (System.Xml.XmlReader reader) [0x00000] in <filename unknown>:0 at System.Configuration.ConfigurationSection.DeserializeSection (System.Xml.XmlReader reader) [0x00000] in <filename unknown>:0 at System.Configuration.Configuration.GetSectionInstance (System.Configuration.SectionInfo config, Boolean createDefaultInstance) [0x00000] in <filename unknown>:0 at System.Configuration.ConfigurationSectionCollection.get_Item (System.String name) [0x00000] in <filename unknown>:0 at System.Configuration.Configuration.GetSection (System.String path) [0x00000] in <filename unknown>:0 at System.Web.Configuration.WebConfigurationManager.GetSection (System.String sectionName, System.String path, System.Web.HttpContext context) [0x00000] in <filename unknown>:0 at System.Web.Configuration.WebConfigurationManager.GetSection (System.String sectionName, System.String path) [0x00000] in <filename unknown>:0 at System.Web.Configuration.WebConfigurationManager.GetWebApplicationSection (System.String sectionName) [0x00000] in <filename unknown>:0 at System.Web.Compilation.BuildManager.get_CompilationConfig () [0x00000] in <filename unknown>:0 at System.Web.Compilation.BuildManager.Build (System.Web.VirtualPath vp) [0x00000] in <filename unknown>:0 at System.Web.Compilation.BuildManager.GetCompiledType (System.Web.VirtualPath virtualPath) [0x00000] in <filename unknown>:0 at System.Web.Compilation.BuildManager.GetCompiledType (System.String virtualPath) [0x00000] in <filename unknown>:0 at System.Web.HttpApplicationFactory.InitType (System.Web.HttpContext context) [0x00000] in <filename unknown>:0

    Read the article

  • Blank black screen with cursor after login -- RHEL5

    - by Sean O.
    I have a RHEL 5 machine here which is a Dell Precision T3500. I'm an Ubuntu guy, but I'm having a heck of a time with this machine. After processing its first security update, we cannot log in via the gdm greeter. A new kernel was installed; then I installed the nVidia drivers for our Quadro NVS 295. I know the X configuration is valid because the gdm greeter does display; however, upon login all we can get is a blank, black screen with a cursor. I thought perhaps our python installation was corrupted but a reinstall via yum has not helped. I have searched & googled extensively for a potential fix for this and can find nothing. Below are outputs from uname, a tail of an error in /var/log/messages, and the Xorg.conf. Can anyone suggest a course of action? [sean@cheetah ~]$ uname -a Linux cheetah.*.* 2.6.18-308.8.1.el5 #1 SMP Fri May 4 16:43:02 EDT 2012 x86_64 x86_64 x86_64 GNU/Linux [sean@cheetah ~]$ sudo tail /var/log/messages Jun 5 15:03:04 cheetah gconfd (sean-4592): Resolved address "xml:readonly:/etc/gconf/gconf.xml.defaults" to a read-only configuration source at position 2 Jun 5 15:03:05 cheetah hcid[3855]: Default passkey agent (:1.8, /org/bluez/applet) registered Jun 5 15:03:05 cheetah pcscd: winscard.c:304:SCardConnect() Reader E-Gate 0 0 Not Found Jun 5 15:03:05 cheetah last message repeated 2 times Jun 5 15:03:06 cheetah gconfd (sean-4592): Resolved address "xml:readwrite:/home/sean/.gconf" to a writable configuration source at position 0 Jun 5 15:03:06 cheetah setroubleshoot: [program.ERROR] exception ImportError: /usr/lib/libatk-1.0.so.0: undefined symbol: g_assertion_message_expr Traceback (most recent call last): File "/usr/bin/sealert", line 952, in ? from setroubleshoot.gui_utils import * File "/usr/lib/python2.4/site-packages/setroubleshoot/gui_utils.py", line 26, in ? import gtk File "/usr/lib64/python2.4/site-packages/gtk-2.0/gtk/__init__.py", line 48, in ? from gtk import _gtk ImportError: /usr/lib/libatk-1.0.so.0: undefined symbol: g_assertion_message_expr Jun 5 15:03:07 cheetah setroubleshoot: [program.ERROR] exception ImportError: /usr/lib/libatk-1.0.so.0: undefined symbol: g_assertion_message_expr Traceback (most recent call last): File "/usr/bin/sealert", line 952, in ? from setroubleshoot.gui_utils import * File "/usr/lib/python2.4/site-packages/setroubleshoot/gui_utils.py", line 26, in ? import gtk File "/usr/lib64/python2.4/site-packages/gtk-2.0/gtk/__init__.py", line 48, in ? from gtk import _gtk ImportError: /usr/lib/libatk-1.0.so.0: undefined symbol: g_assertion_message_expr Jun 5 15:03:08 cheetah pcscd: winscard.c:304:SCardConnect() Reader E-Gate 0 0 Not Found Jun 5 15:07:01 cheetah ntpd[4114]: synchronized to 64.16.211.38, stratum 3 Jun 5 15:07:01 cheetah ntpd[4114]: kernel time sync enabled 0001 [sean@cheetah ~]$ cat /etc/X11/xorg.conf # nvidia-xconfig: X configuration file generated by nvidia-xconfig # nvidia-xconfig: version 295.53 ([email protected]) Sat May 12 00:34:20 PDT 2012 # Xorg configuration created by system-config-display Section "ServerLayout" Identifier "single head configuration" Screen 0 "Screen0" 0 0 InputDevice "Mouse0" "CorePointer" InputDevice "Keyboard0" "CoreKeyboard" EndSection Section "InputDevice" # generated from default Identifier "Mouse0" Driver "mouse" Option "Protocol" "auto" Option "Device" "/dev/input/mice" Option "Emulate3Buttons" "no" Option "ZAxisMapping" "4 5" EndSection Section "InputDevice" Identifier "Keyboard0" Driver "kbd" Option "XkbModel" "pc105" Option "XkbLayout" "us" EndSection Section "Monitor" ### Comment all HorizSync and VertSync values to use DDC: ### Comment all HorizSync and VertSync values to use DDC: Identifier "Monitor0" ModelName "LCD Panel 1600x1200" HorizSync 31.5 - 74.7 VertRefresh 56.0 - 65.0 Option "dpms" EndSection Section "Device" Identifier "Videocard0" Driver "nvidia" EndSection Section "Screen" Identifier "Screen0" Device "Videocard0" Monitor "Monitor0" DefaultDepth 24 SubSection "Display" Viewport 0 0 Depth 24 EndSubSection EndSection

    Read the article

  • OS X Snow Leopard 10.6 Refuses to Load Websites the first time intermittently

    - by Brandon
    Many times when I am browsing the web, Snow Leopard will sit and load a site for 20 seconds or more, until it times out and says it cannot be displayed. If I refresh, it loads RIGHT away, every time. The issue is intermittent but happens from once every couple of days to a few times a day. So the long and short of it is this: Aluminum MacBook (Non-Pro) 2.4GHz Core2Duo, 4GB DDR3 I am using 10.6.6 but I have had this issue since 10.6.0 It happens in Firefox, Chrome, and Safari I have flushed my DNS (using the command 'blablabla flush') I am using custom DNS servers which I hoped would fix it but it had no effect* I am running Apache currently but haven't been for most of the time I've reformatted multiple times, always experiencing the issue I am on Cox cable internet, with a Motorola Surfboard & a Belkin F6D4230-4 v1 (Pre?) N wireless router. I've put the router in G only & N only & G+N to no effect It seems to be domain dependant as I can sometimes load the Google cache right away, and sometimes other sites will load but Google will refuse My Powerbook G4 with Leopard, other Windows XP laptops, & my wired Win7 desktop do not suffer from the issue. *I recently started using these to escape the awful Cox redirect page on timeouts I'm almost positive the issue has happened on other networks but I can't recall a specific instance (I have a terrible memory). The problem is intermittent and fixable enough (I just have to wait until it times out and hit refresh one time) but incredibly annoying since I'm constantly reading documentation from a large variety of sites. EDIT: To clarify, this happens with ALL sites, not only specific sites. I haven't been able to detect any pattern to the failures, but one day Google.com will refuse to load while reddit.com will, and the next day vice versa. Keep in mind that waiting for a timeout and hitting refresh loads the page right away, every time. If I don't wait for the timeout, opening more links, hitting refresh, and clicking the link a billion times have no effect. It seems to be domain neutral, affecting sites seemingly at random. It doesn't seem to have anything to do with connection inactivity either, because I will be SSHed into different servers, uploading files, browsing, downloading, etc, and it will just quit loading Jquery.com (for example) until I sit and wait for a timeout. /EDIT This is my last resort. Please, someone, tell me what is happening. Thank you.

    Read the article

  • Getting Classic ASP to work in .js files under IIS 7

    - by Abdullah Ahmed
    I am moving a clients classic asp webapp to a new IIS7 based server. The site contains some .js files which have javascript but also classic asp in <% % tags which contains a bunch of conditional statements designed to spit out pieces of javascript based on session state variables. Here's a brief example of what the file could be like.... var arrHOFFSET = -1; var arrLeft ="<"; var arrRight = ">"; <% If ((Session("dashInv") = "True") And ((Session("systemLevelStaff") = "4") Or (Session("systemLevelCompany") = "4"))) Then %> addMainItem("/MgmtTools/WelcomeInventory.asp?wherefrom=salesMan","",81,"center","","",0,0,"","","","",""); <% Else %> <% If (Session("dashInv") = "False") And ((Session("systemLevelStaff") = "4") Or (Session("systemLevelCompany") = "4")) Then %> <% Else %> addMainItem("/calendar/welcome.asp","",81,"center","","",0,0,"","","","",""); <% End If %> <% End If %> defineSubmenuProperties(135,"center","center",-3,0,"","","","","","",""); Currently this file (named custom.js for example) will start throwing js errors, because the server doesnt seem to recognize the asp code in it and therefore does not parse it. I know I need to somehow specify that a .js file should also be treated like an .asp file and run through parsing it. However I am not sure how to go about doing this. Here is what I've tried so far... Under the Server node in IIS under HANDLER MAPPINGS I created a new Script Map with the following settings. Request Path: *.js Executable: C:\Windows\System32\inetsrv\asp.dll Name: ASPClassicInJSFiles Mapping: Invoke Handler only if request is mapped to : File Verbs: All verbs Access: Script I also created a similar handler under the site node itself. Under MIME Types .js is defined as application/x-javascript None of these work. If I simply rename the file to have .asp extension then things work, however this app is poorly coded and has literally 100's of files with the .js files included in them under various names and locations, so rename, search and replace is the last option I have.

    Read the article

  • .htaccess working on remote server but does not work on localhost. Getting 404 errors on localhost

    - by Afsheen Khosravian
    MY PROBLEM: When I visit localhost the site does not work. It shows some text from the site but it seems the server can not locate any other files. Here is a snippet of the errors from firebug: "NetworkError: 404 Not Found - localhost/css/popup.css" "NetworkError: 404 Not Found - localhost/css/style.css" "NetworkError: 404 Not Found - localhost/css/player.css" "NetworkError: 404 Not Found - localhost/css/ui-lightness/jquery-ui-1.8.11.custom.css" "NetworkError: 404 Not Found - localhost/js/jquery.js" It seems my server is looking for the files in the wrong places. For example, localhost/css/popup.css is actually located at localhost/app/webroot/css/popup.css. I have my site setup on a remote server with the same exact configurations and it works perfectly fine. I am just having this issue trying to run the site on my laptop at localhost. I edited my VirtualHosts file DocumentRoot and to /home/user/public_html/site.com/public/app/webroot/ and this reduces some errors but I feel that this is wrong and sort of hacking it since I didn't use these setting on my production server which works. The last note I want to make is that the website uses dynamic URLs. I dont know if that has anything to do with it. For example, on the production server the URLS are: site.com/#hello/12321. HERES WHAT I AM WORKING WITH: I have a LAMP server setup on my laptop which runs on Ubuntu 11.10. I have enabled mod_rewrite: sudo a2enmod rewrite Then I edited my Virtual Hosts file: <VirtualHost *:80> ServerName localhost DirectoryIndex index.php DocumentRoot /home/user/public_html/site.com/public <Directory /home/user/public_html/site.com/public/> Options Indexes FollowSymLinks MultiViews AllowOverride All Order allow,deny allow from all </Directory> </VirtualHost> Then I restarted apache. My website is using cakePHP. This is the directory structure of the website: "/home/user/public_html/site.com/public" contains: index.php app cake plugins vendors These are my .htaccess files: /home/user/public_html/site.com/public/app/.htaccess: <IfModule mod_rewrite.c> RewriteEngine on RewriteRule ^$ webroot/ [L] RewriteRule (.*) webroot/$1 [L] </IfModule> /home/user/public_html/site.com/public/app/webroot/.htaccess: <IfModule mod_rewrite.c> RewriteEngine On RewriteCond %{REQUEST_FILENAME} !-d RewriteCond %{REQUEST_FILENAME} !-f RewriteRule ^(.*)$ index.php?url=$1 [QSA,L] </IfModule>

    Read the article

  • Nagios3: Conditional operators for service checks?

    - by Dave
    I'm trying to setup Nagios to monitor my various using hostgroups to define 'machine roles', against which I run services to check the machines by role. However, I'd like to use conditional operators that would enable me to run the service check against an intersection of two host groups, rather than their unions... i.e. using &&, ||, or () operators. For example, imagine I have the following servers: www-eu: Linux WWW (Apache) server, in the EU www-us: Windows WWW (IIS) server, in the US (West coast) ftp-eu: Linux FTP server, in the EU ftp-us: Windows FTP server, in the US I would want to create the following host groups: US-Servers: www-us, ftp-us EU-Servers: www-eu, ftp-eu WWW-Servers: www-us, www-eu FTP-Servers: ftp-us, ftp-eu Now say I'm interested in checking the HTTP response time for my web servers. Then let's say this particular Nagios service is running from the US (West Coast), and that I have a command called *check_http_response_time*. This command will check the responsiveness of the HTTP server, which I can provide an argument which defines the max response time before raising critical. My command might look like: check_http_response_time $HOSTNAME$ 50 Now traditionally, I can run my checks by specifying a list of host or hostgroups. define service{ use local-service hostgroup_name WWW-Servers # Servers = www-us, www-eu servicegroups WWW Checks service_description Check HTTP Response Time check_command check_http_response_time!50 } However, with the above service definition, given my Nagios service is in US West, I could reasonably expect that my EU server will return critical. Really, I want different thresholds for each region (50 for US West, 200 for EU.) I would have to permutate my service for each host and set their custom threshold, or alternatively permutate out my service groups by role & region (i.e. WWW-Servers-EU), and run my specific thresholds against those. Though the latter is better, both are much messier than I'd like... What I would love, and what this post is asking for, is a way to use hostgroups to perform an intersection using conditional logic, rather than a simple union. It might look like: define service{ use local-service hostgroup_name WWW-Servers && US-Servers servicegroups WWW Checks service_description Check HTTP Response Time check_command check_http_response_time!50 } It then would run the check only against servers that are in both WWW-Servers and US-Servers, in my example, just www-us. The benefits of such a feature would be significant for Nagios services configured for large-scale. Is this feature available? If it isn't, will it be available in the future? Is there an alternative way to accomplish this given the most recent Nagios version? Any tips/suggestions are most appreciated! Dave

    Read the article

  • How clean is deleting a computer object?

    - by Kevin
    Though quite skilled at software development, I'm a novice when it comes to Active Directory. I've noticed that AD seems to have a lot of stuff buried in the directory and schema which does not appear superficially when using simplified tools such as Active Directory Users and Computers. It kind of feels like the Windows registry, where COM classes have all kinds of intertwined references, many of which are purely by GUID, such that it's not enough to just search for anything referencing "GadgetXyz" by name in order to cleanly remove GadgetXyz. This occasionally leads to the uneasy feeling that I may have useless garbage building up in there which I have no idea how to weed out. For instance, I made the mistake a while back of trying to rename a DC, figuring I could just do it in the usual manner from Control Panel. I found references to the old name buried all over the place which made it impossible to reuse that name without considerable manual cleanup. Even long after I got it all working, I've stumbled upon the old name hidden away in LDAP. (There were no other DCs left in the picture at that time so I don't think it was a tombstone issue.) More specifically, I'm worried about the case of just outright deleting a computer from AD. I understand the cleanest way to do it is to log into the computer itself and tell it to leave the domain. (As an aside, doing this in Windows 8 seems to only disable the computer object and not delete it outright!) My concern is cases where this is not possible, for instance because it was on an already-deleted VM image. I can simply go into Active Directory Users and Computers, find the computer object, click it, and press Delete, and it seems to go away. My question is, is it totally, totally gone, or could this leave hanging references in any Active Directory nook or cranny I won't know to look in? (Excluding of course the expected tombstone records which expire after a set time.) If so, is there any good way to clean up the mess? Thank you for any insight! Kevin ps., It was over a year ago so I don't remember the exact details, but here's the gist of the DC renaming issue. I started with a single 2008 DC named ABC in a physical machine and wanted to end up instead with a DC of the same name running in a vSphere VM. Not wanting to mess with imaging the physical machine, my plan instead was: Rename ABC to XYZ. Fresh install 2008 on a VM, name it ABC, and join it to the domain. (I may have done the latter in the same step as promoting to DC; I don't recall.) dcpromo the new ABC as a 2nd DC, including GC. Make sure the new ABC replicated correctly from XYZ and then transfer the FSMO roles from XYZ to it. Once everything was confirmed to work with the new ABC alone, demote XYZ, remove the AD role, and remove it from the domain. Eventually I managed to do this but it was a much bumpier ride than expected. In particular, I got errors trying to join the new ABC to the domain. These included "The pre-windows 2000 name is already in use" and "No mapping between account names and security IDs was done." I eventually found that the computer object for XYZ had attributes that still referred to it as ABC. Among these were servicePrincipalName, msDS-AdditionalDnsHostName, and msDS-AdditionalSamAccountName. The latter I could not edit via Attribute Editor and instead had to run this against XYZ: NETDOM computername <simple-name> /add:<FQDN> There were some other hitches I don't remember exactly.

    Read the article

  • MySQL Execution Time Spikes

    - by Brett
    I am having issues with MySQL all of the sudden today. Details: OS: CentOS release 5.7 Server type: Parallels virtuozzo container running on mediatemple DV 4.0 package Average total memory usage: <500mb Total memory usage allowed: 1gb (part of shared pool for emergency only, users are only guaranteed 500mb) Processor: 1ghz Main database sizes with most usage: 275mb & 107mb server stack: nginx 1.0.10, mysql 5.1.54, php 5.3.8 with php-fpm innodb_buffer_pool_size=100M php-fpm max children: 5 Webapps: custom php-based sites, magento & drupal slow query timeout is set to 1 second Steps I completed towards diagnosis: Cannot restart container yet - I will try later tonight when our domestic traffic has dropped Enabled mysql and php-fpm slowlog. Found functions that did DB queries in php-fpm slowlog were taking over 1s to complete at times Found some simple queries in mysql slowlog taking well over 1s to complete that should take less than 1s. Most interesting - execution time seems to spike at times. A query will take .2s a couple times, then one time it will take 8s to run the same query. These results were verified by running raw SQL queries through mysql command line. Top does not reveal anything too interesting Only resource related thing i can see is load averages much higher than normal Up until today, mysql has been fine, there have been no major changes to the db since yesterday. Sometimes things are so bad, I am seeing bad gateway errors after 60s of execution time. Innodb is doing on average 300-1400 reads/sec. Mysql is doing 3-10 queries/sec slow query count in 2 hours uptime is 171 (with slow timeout at 1 second) Tried restarting mysql, nginx, php-fpm multiple times For example: UPDATE `catalogsearch_query` SET `query_text` = 'EW 90', `num_results` = '7532', `popularity` = '99180', `redirect` = NULL, `synonym_for` = NULL, `store_id` = '1', `display_in_terms` = '1', `is_active` = '1', `is_processed` = '1', `updated_at` = '2012-05-08 21:38:31' WHERE (query_id='31'); This query took 17sec to complete one time, rest of the time around .079 sec. But varies, sometimes 1sec, sometimes .004 sec. This is running the same query, over and over with a couple seconds time in between each. Most tables are innodb, and sometimes I noticed the lock time taking 90% of the query execution time, but most of the time lock time is insignificant. Any idea what's going on here?

    Read the article

  • Why is IIS Anonymous authentication being used with administrative UNC drive access?

    - by Mark Lindell
    My account is local administrator on my machine. If I try to browse to a non-existent drive letter on my own box using a UNC path name: \mymachine\x$ my account would get locked out. I would also get the following warning (Event ID 100, Type “Warning”) 5 times under the “System” group in Event Viewer on my box: The server was unable to logon the Windows NT account 'ourdomain\myaccount' due to the following error: Logon failure: unknown user name or bad password. I would also get the following warning 3 times: The server was unable to logon the Windows NT account 'ourdomain\myaccount' due to the following error: The referenced account is currently locked out and may not be logged on to. On the domain controller, Event ID 680 of type “Failure Audit” would appear 4 times under the “Security” group in Event Viewer: Logon attempt by: MICROSOFT_AUTHENTICATION_PACKAGE_V1_0 Logon account: myaccount Followed by Event ID 644: User Account Locked Out: Target Account Name: myaccount Target Account ID: OURDOMAIN\myaccount Caller Machine Name: MYMACHINE Caller User Name: STAN$ Caller Domain: OURDOMAIN Caller Logon ID: (0x0,0x3E7) Followed by another 4 errors having Event ID 680. Strangely, every time I tried to browse to the UNC path I would be prompted for a user name and password, the above errors would be written to the log, and my account would be locked out. When I hit “Cancel” in response to the user name/password prompt, the following message box would display: Windows cannot find \mymachine\x$. Check the spelling and try again, or try searching for the item by clicking the Start button and then clicking Search. I checked with others in the group using XP and they only got the above message box when browsing to a “bad” drive letter on their box. No one else was prompted for a user name/password and then locked out. So, every time I tried to browse to the “bad” drive letter, behind the scenes XP was trying to login 8 times using bad credentials (or, at least a bad password as the login was correct), causing my account to get locked out on the 4th try. Interestingly, If I tried browsing to a “good” drive such as “c$” it would work fine. As a test, I tried logging on to my box as a different login and browsing the “bad” UNC path. Strangely, my “ourdomain\myaccount” account was getting locked out – not the one I was logged in as! I was totally confused as to why the credentials for the other login were being passed. After much Googling, I found a link referring to some IIS settings I was vaguely familiar with from the past but could not see how they would affect this issue. It was related to the IIS directory security setting “Anonymous access and authentication control” located under: Control Panel/Administrative Tools/Computer Management/Services and Applications/Internet Information Services/Web Sites/Default Web Site/Properties/Directory Security/Anonymous access and authentication control/Edit/Password I found no indication while scouring the Internet that this property was related to my UNC problem. But, I did notice that this property was set to my domain user name and password. And, my password did age recently but I had not reset the password accordingly for this property. Sure enough, keying in the new password corrected the problem. I was no longer prompted for a user name/password when browsing the UNC path and the account lock-outs ceased. Now, a couple of questions: Why would an IIS setting affect the browsing of a UNC path on a local box? Why had I not encountered this problem before? My password has aged several times and I’ve never encountered this problem. And, I can’t remember the last time I updated the “Anonymous access” IIS password it’s been so long. I’ve run the script after a password reset before and never had my account locked-out due to the UNC problem (the script accesses UNC paths as a normal part of its processing). Windows Update did install “Cumulative Security Update for Internet Explorer 7 for Windows XP (KB972260)” on my box on 7/29/2009. I wonder if this is responsible.

    Read the article

  • OpenVPN not sending traffic to internet?

    - by coleifer
    I've set up openvpn on my pi and am running into a small issue. I can connect to the VPN server and ping it just fine, and I can also connect to other machines on my local network. However I am unable, when connected to the VPN, to reach the outside world (either by name lookup or IP). here are the details: On the server the tun0 interface: tun0: flags=4305<UP,POINTOPOINT,RUNNING,NOARP,MULTICAST> mtu 1500 inet 10.8.0.1 netmask 255.255.255.255 destination 10.8.0.2 unspec 00-00-00-00-00-00-00-00-00-00-00-00-00-00-00-00 txqueuelen 100 (UNSPEC) RX packets 0 bytes 0 (0.0 B) RX errors 0 dropped 0 overruns 0 frame 0 TX packets 0 bytes 0 (0.0 B) TX errors 0 dropped 0 overruns 0 carrier 0 collisions 0 I can ping it just fine: # ping -c 3 10.8.0.1 PING 10.8.0.1 (10.8.0.1) 56(84) bytes of data. 64 bytes from 10.8.0.1: icmp_seq=1 ttl=64 time=0.159 ms 64 bytes from 10.8.0.1: icmp_seq=2 ttl=64 time=0.155 ms 64 bytes from 10.8.0.1: icmp_seq=3 ttl=64 time=0.156 ms --- 10.8.0.1 ping statistics --- 3 packets transmitted, 3 received, 0% packet loss, time 2002ms Routing table # ip route show default via 192.168.1.1 dev eth0 metric 204 10.8.0.0/24 via 10.8.0.2 dev tun0 10.8.0.2 dev tun0 proto kernel scope link src 10.8.0.1 192.168.1.0/24 dev eth0 proto kernel scope link src 192.168.1.6 metric 204 I also have ip traffic forwarding: net.ipv4.ip_forward = 1 I do not have any custom iptables rules (that I'm aware of). On the client, I can connect to the VPN. Here is my tun0: tun0: flags=4305<UP,POINTOPOINT,RUNNING,NOARP,MULTICAST> mtu 1500 inet 10.8.0.6 netmask 255.255.255.255 destination 10.8.0.5 unspec 00-00-00-00-00-00-00-00-00-00-00-00-00-00-00-00 txqueuelen 100 (UNSPEC) RX packets 0 bytes 0 (0.0 B) RX errors 0 dropped 0 overruns 0 frame 0 TX packets 21 bytes 1527 (1.4 KiB) TX errors 0 dropped 0 overruns 0 carrier 0 collisions 0 And on the client I can ping it: sudo ping -c 3 10.8.0.6 PING 10.8.0.6 (10.8.0.6) 56(84) bytes of data. 64 bytes from 10.8.0.6: icmp_seq=1 ttl=64 time=0.035 ms 64 bytes from 10.8.0.6: icmp_seq=2 ttl=64 time=0.026 ms 64 bytes from 10.8.0.6: icmp_seq=3 ttl=64 time=0.032 ms --- 10.8.0.6 ping statistics --- 3 packets transmitted, 3 received, 0% packet loss, time 1998ms rtt min/avg/max/mdev = 0.026/0.031/0.035/0.003 ms I can ssh from the client into another server on my LAN (192.168.1.x), however I cannot reach anything outside my LAN. Here's some of the server logs at the bottom of this gist: https://gist.github.com/coleifer/6ef95c3008f130249933/edit I am frankly out of ideas! I don't think it's my client because both my laptop and my phone (which has an openvpn client) exhibit the same behavior. I had OpenVPN installed on this pi before using debian and it worked, so I don't think it's my router but of course anything is possible.

    Read the article

  • Odd squid transparent redirect behavior

    - by EMiller
    This is the first time I've set up squid. It's running a redirect script that does some text search/replace on html pages, and then saves them to a location on the same machine on the nginx path - then issues the redirect to that URL (it's an art project :D). The relevant lines in squid.conf are http_port 3128 transparent redirect_program /etc/squid/jefferson_redirect.py The jefferson_redirect.py script is based on this script: http://gofedora.com/how-to-write-custom-redirector-rewritor-plugin-squid-python/ The issue: I'm getting strange http redirect behavior. For example, here is the normal request/response from a PHP script that issues a header("Location:"); - a 302 redirect: http://redirector.mysite.com/?unicmd=g+yreka GET /?unicmd=g+yreka HTTP/1.1 Host: redirector.mysite.com User-Agent: Mozilla/5.0 (X11; U; Linux x86_64; en-US; rv:1.9.1.9) Gecko/20100330 Fedora/3.5.9-1.fc12 Firefox/3.5.9 Accept: text/html,application/xhtml+xml,application/xml;q=0.9,*/*;q=0.8 Accept-Language: en-us,en;q=0.5 Accept-Encoding: gzip,deflate Accept-Charset: ISO-8859-1,utf-8;q=0.7,*;q=0.7 Keep-Alive: 300 Connection: keep-alive HTTP/1.1 302 Found Date: Tue, 13 Apr 2010 05:15:43 GMT Server: Apache X-Powered-By: PHP/5.2.11 Location: http://www.google.com/search?q=yreka Content-Type: text/html Vary: User-Agent,Accept-Encoding Content-Encoding: gzip Content-Length: 2108 Keep-Alive: timeout=3, max=100 Connection: Keep-Alive Here's what it looks like when running through the squid proxy (note that "redirector.mysite.com" is not the site running squid or nginx): http://redirector.mysite.com/?unicmd=g+yreka GET /?unicmd=g+yreka HTTP/1.1 Host: redirector.mysite.com User-Agent: Mozilla/5.0 (X11; U; Linux x86_64; en-US; rv:1.9.1.9) Gecko/20100330 Fedora/3.5.9-1.fc12 Firefox/3.5.9 Accept: text/html,application/xhtml+xml,application/xml;q=0.9,*/*;q=0.8 Accept-Language: en-us,en;q=0.5 Accept-Encoding: gzip,deflate Accept-Charset: ISO-8859-1,utf-8;q=0.7,*;q=0.7 Keep-Alive: 300 Proxy-Connection: keep-alive If-Modified-Since: Tue, 13 Apr 2010 05:21:02 GMT HTTP/1.0 200 OK Server: nginx/0.7.62 Date: Tue, 13 Apr 2010 05:21:10 GMT Content-Type: text/html Content-Length: 17865 Last-Modified: Tue, 13 Apr 2010 05:21:10 GMT Accept-Ranges: bytes X-Cache: MISS from jefferson X-Cache-Lookup: HIT from jefferson:3128 Via: 1.1 jefferson:3128 (squid/2.7.STABLE6) Connection: keep-alive Proxy-Connection: keep-alive It is basically working - but the URL http://redirector.mysite.com/?unicmd=g+yreka remains unchanged, while displaying the google page (mostly broken as it's using URLs relative to redirector.mysite.com) I've experienced a similar thing with google results pages: when clicking to another page from google, I get a google URL, with the other site's content. Sorry for the long post - many thanks if you've read this far! Any ideas?

    Read the article

  • Remote installing an msi on citrix servers using WMI

    - by capn
    OK, I'm a C# programmer that is trying to streamline the deployment of a custom windows form app I inherited and built an installer for with WiX (this app will need to be reinstalled regularly as I'm making changes to it). I'm not really used to admin type things (or vbs, or WMI, or terminal servers, or Citrix, and even WiX and MSI are not things I usually deal with) but so far I put together some vbs and have an end goal in mind. The msi does work, and I've installed it from the mapped O: drive on my dev machine and while RDP'd to a citrix machine. End Goal: Deploy code written on my dev machine and compiled into an MSI (that I can improve upon within the confines of WiX and whatever the Windows Installer Engine allows) to the cluster of Citrix machines my users have access to. What am I missing in my script to get the MSI to execute on the remote machines? Layout: Machine A is my dev machine, and has the vbs script and the msi file (XP SP3) Machines C1 - C6 are the Citrix Servers that need the application installed them via the msi (Server 2003 R2 SP2) There is also optionally a shared network resource that all the machines can access. Script: 'Set WMI Constants Const wbemImpersonationLevelImpersonate = 3 Const wbemAuthenticationLevelPktPrivacy = 6 'Set whether this is installing to the debug Citrix Servers Const isDebug = true 'Set MSI location 'Network location yields error 1619 (This installation package could not be opened.) msiLocation = "\\255.255.255.255\odrive\Citrix Deployment\Setup.msi" 'Directory on machine A yields error 3 (file not found) 'msiLocation = "C:\Temp\Deploy\Setup.msi" 'Mapped network drive (on both machines) yield error 3 (file not found) 'msiLocation = "O:\Citrix Deployment\Setup.msi" 'Set login information strDomain = "MyDomain" Wscript.StdOut.Write "user name:" strUser = Wscript.StdIn.ReadLine Set objPassword = CreateObject("ScriptPW.Password") Wscript.StdOut.Write "password:" strPassword = objPassword.GetPassword() 'Names of Citrix Servers Dim citrixServerArray If isDebug Then citrixServerArray = array("C4") Else 'citrixServerArray = array("C1","C2","C3","C5","C6") End If 'Loop through each Citrix Server For Each citrixServer in citrixServerArray 'Login to remote computer Set objLocator = CreateObject("WbemScripting.SWbemLocator") Set objWMIService = objLocator.ConnectServer(citrixServer, _ "root\cimv2", _ strUser, _ strPassword, _ "MS_409", _ "ntlmdomain:" + strDomain) 'Set Remote Impersonation level objWMIService.Security_.ImpersonationLevel = wbemImpersonationLevelImpersonate objWMIService.Security_.AuthenticationLevel = wbemAuthenticationLevelPktPrivacy 'Reference to a process on the machine Dim objProcess : Set objProcess = objWMIService.Get("Win32_Process") 'Change user to install for terminal services errReturn = objProcess.Create _ ("cmd.exe /c change user /install", Null, Null, intProcessID) WScript.Echo errReturn 'Install MSI here 'Reference to a product on the machine Set objSoftware = objWMIService.Get("Win32_Product") 'All users set in option parameter, I'm led to believe that the third parameter is actually ignored 'http://www.webmasterkb.com/Uwe/Forum.aspx/vbscript/2433/Installing-programs-with-VbScript errReturn = objSoftware.Install(msiLocation,"ALLUSERS=2 REBOOT=ReallySuppress",True) Wscript.Echo errReturn 'Change user back to execute errReturn = objProcess.Create _ ("cmd.exe /c change user /execute", Null, Null, intProcessID) WScript.Echo errReturn Next I also tried using this to install, it doesn't return an error code, but doesn't install the msi either, and it makes me wonder if the change user /install command is even really working. errReturn = objProcess.Create _ ("cmd.exe /c msiexec /i ""O:\Citrix Deployment\Setup.msi"" /quiet") Wscript.Echo errReturn

    Read the article

  • Add a small RAID card? Will it help overall stability and performance of my nine hard drives?

    - by Ray
    Hi, Will I get any extra genuine added performance and RAID stability if I insert a basic RAID card into a PCI-E x1 slot? I am considering the Adaptec 1220SA - 2 port SATA , pci-express (1x) , raid 0/1. Ok it only supports two SATA drives. Purpose is to help support the eight internal hard drives (1TB each), a DVD drive and an external e-SATA connected 2TB hard drive - by dealing with two of the internal hard drives. My current configuration of eight internal 1TB Barracuda (7200.12) SATA hard drives, one external 2TB SATA Western Digital Green Drive (e-SATA) and one DVD drive can already be supported by the Intel P55 & JMicron controllers on the ASUS motherboard : the Intel P55 (controls six HDD; configured as three x RAID 1), and the JMicron (controls two HDD as one RAID 1, as well as the DVD drive and the external SATA drive via the motherboard's e-SATA port (controlled by the JMicron)). Bigger picture details : I have an ASUS motherboard designed for the LGA1156 type processor and it includes the Intel P55 Express Chipset and JMicron. I am using the Intel Core i7-870 processor, and have 8GB DDR3 (1333) memory (four x 2GB Corsair DIMMs). Enough overall power. The power supply is more than sufficicient for the system. Corsair AX850. The system will never need the full 850 watts (future : second graphics card). The RAID card would provide hardware RAID 1 for two of the eight intrnal drives. It would either reduce the load on : the Intel P55 firmware RAID support, or replace the JMicron controller's RAID 1 set. I am busy installing the above configuration using Windows 7 Ultimate 64-bit as the OS. The RAID card is a last minute addition to the plan. Is it worth spending the extra R700 - R900 on the Adaptec 1220SA, or equivalent RAID card? I cannot afford to spend yet another R2000 - R3000 on a RAID card that would support many SATA2 hard drives, with a better RAID, example the RAID 5. My Issue & assumption : I am trusting that the Intel P55 chipset can properly handle six drives, configured as three * RAID 1. I am assuming that the JMicron can handle, using its RED SATA ports, one RAID-1 (two HDDs). The DVD drive connects to the JMicron optical SATA port 1 (white port 1). White port 2 is not used. The e-SATA connection is from the JMicron straight to, and through the motherboard - to an on-board (rear panel) e-SATA port. Am I being a little hopeful in only using the on-board Intel P55 and the JMicron? Is it a waste of money to install a RAID card that handles two SATA2 drives? OR Is it wisdom to take the pressure a little off the Intel P55? Obviously I am interested in data security, hence RAID 1, not RAID Zero. RAID 5 would be nice. The CPU, Intel Core i7-870 will provide the clout. Context to nine drives : I am using virtualisation with Windows 7 Ultimate. Bootable VMs. The operating system gets a mirror. Loaded apps gets a mirror. The current design data is kept in another mirror and Another mirror is back-up one and / or VM territory. Then the external 2TB drive (via e-SATA) is the next layer of data security and then finally, I use off-site data security. Thanks.

    Read the article

  • Dual head setup for Ubuntu 10.04.1 and Windows XP Pro with same hardware configuration

    - by mejpark
    I have a Dell OptiPlex 360 workstation at work, with 2 x ATI RV280 [Radeon 9200 PRO] graphics cards installed, which are attached to two identical 19" HII flat panel monitors. I'm using the open source Radeon driver with Ubuntu, and the proprietary drivers with Windows. The good news is that dual head configuration works for both OSes. The bad news is, I have to use a different hardware configuration for each OS to achieve this. Hardware config #1: Dual monitors work for Windows XP Pro like this: First display -> external VGA port Second display -> DVI input on gfx card Hardware config #2: Dual monitors work for Ubuntu 10.04.1 like this: First display -> VGA port on gfx card Second display -> DVI input on gfx card I connected up the displays according to Config #2 and booted up Windows, which resulted in a mirror image on both screens. I was unable to login, as the login box was not visible. I unplugged the VGA lead from gfx card and plugged it into the external VGA port (Config #1) - Windows dual head works again, but the VGA-connected screen is not recognised by Ubuntu and remains in standby mode. Is it possible to configure a dual head setup for Ubuntu using Config #1, or am I missing something? I tried setting up dual monitors using Config #1, this morning which didn't work. By default, there is no xorg.conf file in Ubuntu 10.04.1, so I generated one using: $ sudo X :2 -configure X.Org X Server 1.7.6 Release Date: 2010-03-17 X Protocol Version 11, Revision 0 Build Operating System: Linux 2.6.24-27-server i686 Ubuntu Current Operating System: Linux harrier 2.6.32-24-generic #42-Ubuntu SMP Fri Aug 20 14:24:04 UTC 2010 i686 Kernel command line: BOOT_IMAGE=/boot/vmlinuz-2.6.32-24-generic root=UUID=a34c1931-98d4-4a34-880c-c227a2936c4a ro quiet splash Build Date: 21 July 2010 12:47:34PM xorg-server 2:1.7.6-2ubuntu7.3 (For technical support please see http://www.ubuntu.com/support) Current version of pixman: 0.16.4 Before reporting problems, check http://wiki.x.org to make sure that you have the latest version. Markers: (--) probed, (**) from config file, (==) default setting, (++) from command line, (!!) notice, (II) informational, (WW) warning, (EE) error, (NI) not implemented, (??) unknown. (==) Log file: "/var/log/Xorg.2.log", Time: Mon Sep 13 10:02:02 2010 List of video drivers: apm ark intel mach64 s3virge trident mga tseng ati nouveau neomagic i740 openchrome voodoo s3 i128 radeon siliconmotion nv ztv vmware v4l chips rendition savage sisusb tdfx geode sis r128 cirrus fbdev vesa (++) Using config file: "/home/michael/xorg.conf.new" (==) Using config directory: "/usr/lib/X11/xorg.conf.d" (II) [KMS] No DRICreatePCIBusID symbol, no kernel modesetting. Xorg detected your mouse at device /dev/input/mice. Please check your config if the mouse is still not operational, as by default Xorg tries to autodetect the protocol. Xorg has configured a multihead system, please check your config. Your xorg.conf file is /home/michael/xorg.conf.new To test the server, run 'X -config /home/michael/xorg.conf.new' ddxSigGiveUp: Closing log $ sudo X -config /home/michael/xorg.conf.new Fatal server error: Server is already active for display 0 If this server is no longer running, remove /tmp/.X0-lock and start again. Please consult the The X.Org Foundation support at http://wiki.x.org for help. ddxSigGiveUp: Closing log I then booted Ubuntu in failsafe mode, dropped into root shell, and executed $ X -config /home/michael/xorg.conf.new again. The screen went blank and turned off, so

    Read the article

  • Workaround with paths in vista

    - by Argiropoulos Stavros
    Not really a programming question but i think quite relative. I have a .exe running in my windows XP PC. This exe needs a file in the same directory to run and has no problem finding it in XP BUT in Vista(I tried this in several machines and works in some of them) fails to run. I'm guessing there is a problem finding the path.The program is written in basic(Yes i know..) I attach the code below. Can you think of any workarounds? Thank you The exe is located in c:\tools Also the program runs in windows console(It starts but then during the execution cannot find a custom file type .TOP made by the creator of the program) ' PROGRAMM TOP11.BAS DEFDBL A-Z CLS LOCATE 1, 1 COLOR 14, 1 FOR i = 1 TO 80 PRINT "±"; NEXT i LOCATE 1, 35: PRINT "?? TOP11 ??" PRINT " €€‚—‚„‘ ’— ‹„’†‘„— ‘’† „”€„€ ’†‘ ‡€€‘‘†‘ ‰€ † „.‚.‘.€. " COLOR 7, 0 PRINT "-------------------------------------------------------------------------------" PRINT INPUT "ƒ?©« «¦¤ ©¬¤«?©«? ¤???... : ", Factor# INPUT "¤¦£ ¨®e¦¬ [.TOP] : ", topfile$ VIEW PRINT 7 TO 25 file1$ = topfile$ + ".TOP" file2$ = topfile$ + ".T_P" file3$ = "Syntel" OPEN file3$ FOR OUTPUT AS #3 PRINT #3, " ‘¬¤«?©«?? ¤??? = " + STR$(Factor#) + " †‹„‹†€: " + DATE$ CLOSE #3 command1$ = "copy" + " " + file1$ + " " + file2$ SHELL command1$ '’¦ ¨®e¦ .TOP ¤« ¨a­«  £ «¤ ?«a?¥ .T_P OPEN file2$ FOR INPUT AS #1 OPEN file1$ FOR OUTPUT AS #2 bb$ = " \\\ \ , ###.#### ###.#### ####.### ##.### " DO LINE INPUT #1, Line$ Line$ = RTRIM$(LTRIM$(Line$)) icode$ = LEFT$(Line$, 1) IF icode$ = "1" THEN Line$ = " " + Line$ PRINT #2, Line$ PRINT Line$ ELSEIF icode$ = "2" THEN Line$ = " " + Line$ PRINT #2, Line$ PRINT Line$ ELSEIF icode$ = "3" THEN Number$ = MID$(Line$, 3, 6) Hangle = VAL(MID$(Line$, 14, 9)) Zangle = VAL(MID$(Line$, 25, 9)) Distance = VAL(MID$(Line$, 36, 9)) Distance = Distance * Factor# Height = VAL(MID$(Line$, 48, 6)) PRINT #2, USING bb$; icode$; Number$; Hangle; Zangle; Distance; Height PRINT USING bb$; icode$; Number$; Hangle; Zangle; Distance; Height ELSE END IF LOOP UNTIL EOF(1) VIEW PRINT CLS LOCATE 1, 1 PRINT " *** ’„‘ ’“ ‚€‹‹€’‘ *** " END

    Read the article

  • Solution to route/proxy SNMP Traps (or Netflow, generic UDP, etc) for network monitoring?

    - by Christopher Cashell
    I'm implementing a network monitoring solution for a very large network (approximately 5000 network devices). We'd like to have all devices on our network send SNMP traps to a single box (technically this will probably be an HA pair of boxes) and then have that box pass the SNMP traps on to the real processing boxes. This will allow us to have multiple back-end boxes handling traps, and to distribute load among those back end boxes. One key feature that we need is the ability to forward the traps to a specific box depending on the source address of the trap. Any suggestions for the best way to handle this? Among the things we've considered are: Using snmptrapd to accept the traps, and have it pass them off to a custom written perl handler script to rewrite the trap and send it to the proper processing box Using some sort of load balancing software running on a Linux box to handle this (having some difficulty finding many load balancing programs that will handle UDP) Using a Load Balancing Appliance (F5, etc) Using IPTables on a Linux box to route the SNMP traps with NATing We've currently implemented and are testing the last solution, with a Linux box with IPTables configured to receive the traps, and then depending on the source address of the trap, rewrite it with a destination nat (DNAT) so the packet gets sent to the proper server. For example: # Range: 10.0.0.0/19 Site: abc01 Destination: foo01 iptables -t nat -A PREROUTING -p udp --dport 162 -s 10.0.0.0/19 -j DNAT --to-destination 10.1.2.3 # Range: 10.0.33.0/21 Site: abc01 Destination: foo01 iptables -t nat -A PREROUTING -p udp --dport 162 -s 10.0.33.0/21 -j DNAT --to-destination 10.1.2.3 # Range: 10.1.0.0/16 Site: xyz01 Destination: bar01 iptables -t nat -A PREROUTING -p udp --dport 162 -s 10.1.0.0/16 -j DNAT --to-destination 10.3.2.1 This should work with excellent efficiency for basic trap routing, but it leaves us completely limited to what we can mach and filter on with IPTables, so we're concerned about flexibility for the future. Another feature that we'd really like, but isn't quite a "must have" is the ability to duplicate or mirror the UDP packets. Being able to take one incoming trap and route it to multiple destinations would be very useful. Has anyone tried any of the possible solutions above for SNMP traps (or Netflow, general UDP, etc) load balancing? Or can anyone think of any other alternatives to solve this?

    Read the article

  • Suspend only works once after full power cycle with ASUS P7P55D-E Pro

    - by John Chadwick
    This one is strange. I can't seem to get suspend working more than once per power cycle. When I say "power cycle," I mean the only way to get one proper suspend is to cut power from the power supply and boot back up cold. After the proper suspend, I get a failed suspend, and after all reboots or cold boots until power is cut, suspends fail. I'm using an ASUS P7P55D-E Pro with a Sandy Bridge Core i7, running on Ubuntu Precise repositories and UEFI. I'm running Nouveau from repository (And Gallium3d compiled from git, but that does not come into this since I can avoid OpenGL and it still happens the same way) with a GTX 285 (nv50.) I had to build a custom kernel (3.3) in order for ACPI 5.0 to be supported and make suspend work at all. I compiled it using the latest Ubuntu kernel's config file with the additional entries set to the default options. All packages are up to date. I know these are relatively exotic settings, but I'm hoping maybe I can get some help anyways. The behavior when suspend fails is strange. Upon a proper suspend, all fans turn off and the only led left on, the power led, is blinking. Upon a failed suspend, 1. USB power remains. 2. The power led stays on solid. 3. All fans seem to still be on. 4. I can hear what I believe is the primary harddrive shutting off. 5. Despite USB power remaining, the USB powered keyboard does not respond to anything, and the indicator leds on it shut off. Pressing the power button does nothing, and of course I have not to date found a way to wake it up. When trouble shooting the first round of issues I got with suspend not too long ago, I ended up building a list of modules to disable upon sleeping. Here's my config file for them: In /etc/pm/config.d/01modules: SUSPEND_MODULES="uhci_hd ehci_hd button" All of my other pm configuration files are stock. In case it's any help, here are my relevant BIOS settings. Thanks.

    Read the article

  • HAProxy causing delay

    - by user1221444
    I am trying to configure HAProxy to do load balancing for a custom webserver I created. Right now I am noticing an increasing delay with HAProxy as the size of the return message increases. For example, I ran four different tests, here are the results: Response 15kb through HAProxy: Avg. response time: .34 secs Transacation rate: 763 trans/sec Throughput: 11.08 MB/sec Response 2kb through HAProxy: Avg. response time: .08 secs Transaction rate: 1171 trans / sec Throughput: 2.51 MB/sec Response 15kb directly to server: Avg. response time: .11 sec Transaction rate: 1046 trans/sec throughput: 15.20 MB/sec Response 2kb directly to server: Avg. Response time: .05 secs Transaction rate: 1158 trans/sec Throughput: 2.48 MB/sec All transactions are HTTP requests. As you can see, there seems to be a much bigger difference between response times for when the response is bigger, than when it is smaller. I understand there will be a slight delay when using HAProxy. Not sure if it matters, but the test itself was run using siege. And during the test there was only one server behind the HAProxy(the same that was used in the direct to server tests). Here is my haproxy.config file: global log 127.0.0.1 local0 log 127.0.0.1 local1 notice maxconn 10000 user haproxy group haproxy daemon #debug defaults log global mode http option httplog option dontlognull retries 3 option redispatch option httpclose maxconn 10000 contimeout 10000 clitimeout 50000 srvtimeout 50000 balance roundrobin stats enable stats uri /stats listen lb1 10.1.10.26:80 maxconn 10000 server app1 10.1.10.200:8080 maxconn 5000 I couldn't find much in terms of options in this file that would help my problem. I have heard suggestions that I may have to adjust a few of my sysctl settings. I could not find a lot of information on this however, most documentation is for Linux 2.4 and 2.6 on the sysctl stuff, I am running 3.2(Ubuntu server 12.04), which seems to auto tuning, so I have no clue what I should or shouldn't be changing. Most settings changes I tried had no effect or a negative effect on performance. Just a notice, this is a very preliminary test, and my hope is that at deployment time, my HAProxy will be able to balance 10k-20k requests/sec to many servers, so if anyone could provide information to help me reach that goal, it would be much appreciated. Thank you very much for any information you can provide. And if you need anymore information from me please let me know, I will get you anything I can.

    Read the article

  • Aliased network interfaces and isc dhcp server

    - by Jonatan
    I have been banging my head on this for a long time now. There are many discussions on the net about this and similar problems, but none of the solutions seems to work for me. I have a Debian server with two ethernet network interfaces. One of them is connected to internet, while the other is connected to my LAN. The LAN network is 10.11.100.0 (netmask 255.255.255.0). We have some custom hardware that use network 10.4.1.0 (netmask 255.255.255.0) and we can't change that. But we need all hosts on 10.11.100.0 to be able to connect to devices on 10.4.1.0. So I added an alias for the LAN network interface so that the Debian server acts as a gateway between 10.11.100.0 and 10.4.1.0. But then the dhcp server stopped working. The log says: No subnet declaration for eth1:0 (no IPv4 addresses). ** Ignoring requests on eth1:0. If this is not what you want, please write a subnet declaration in your dhcpd.conf file for the network segment to which interface eth1:1 is attached. ** No subnet declaration for eth1:1 (no IPv4 addresses). ** Ignoring requests on eth1:1. If this is not what you want, please write a subnet declaration in your dhcpd.conf file for the network segment to which interface eth1:1 is attached. ** I had another server before, also running Debian but with the older dhcp3 server, and it worked without any problems. I've tried everything I can think of in dhcpd.conf etc, and I've also compared with the working configuration in the old server. The dhcp server need only handle devices on 10.11.100.0. Any hints? Here's all relevant config files: /etc/default/isc-dhcp-server INTERFACES="eth1" /etc/network/interfaces (I've left out eth0, that connects to the Internet, since there is no problem with that.) auto eth1:0 iface eth1:0 inet static address 10.11.100.202 netmask 255.255.255.0 auto eth1:1 iface eth1:1 inet static address 10.4.1.248 netmask 255.255.255.0 /etc/dhcp/dhcpd.conf ddns-update-style none; option domain-name "???.com"; option domain-name-servers ?.?.?.?; default-lease-time 86400; max-lease-time 604800; authorative; subnet 10.11.100.0 netmask 255.255.255.0 { option subnet-mask 255.255.255.0; pool { range 10.11.100.50 10.11.100.99; } option routers 10.11.100.102; } I have tried to add shared-network etc, but didn't manage to get that to work. I get the same error message no matter what...

    Read the article

  • Interop.Outlook.UserProperties.Add causing problem during connection time

    - by aanataliya
    Hi All, I have created a plug-in for outlook. Plug-in has only below code. private void OnNewOutlookInspector(Outlook.Inspector OutlookInsptr) { Outlook.MailItem MlItem = (Outlook.MailItem)OutlookInsptr.CurrentItem; //if I remove below line. Everything is working fine. MlItem.UserProperties.Add("INSPINIT", Outlook.OlUserPropertyType.olText , true , true ).Value = "1"; } public void OnConnection(object application, Extensibility.ext_ConnectMode connectMode, object addInInst, ref System.Array custom) { applicationObject = application; addInInstance = addInInst; MessageBox.Show("in connection new 2"); OutlkApp = (Outlook.Application)application; OutlkInsptrs = OutlkApp.Inspectors; OutlkInsptrs.NewInspector += new Outlook.InspectorsEvents_NewInspectorEventHandler(OnNewOutlookInspector); } Problem I am facing is, When I send HTML mail while plug-in is enabled, receiving end it is being received as a plain text. Below is the mail content along with the header and body at recieving end. x-sender: [email protected] x-receiver: [email protected] Received: from blr-s-07.pointcrossblr.com ([192.168.1.107]) by blr-ws-134.pointcrossblr.com with Microsoft SMTPSVC(6.0.2600.5949); Wed, 22 Dec 2010 17:11:02 +0530 Received: from blrws134 ([192.168.1.175]) by blr-s-07.pointcrossblr.com with Microsoft SMTPSVC(6.0.3790.4675); Wed, 22 Dec 2010 17:11:02 +0530 From: "Ashif Nataliya" <[email protected]> To: <[email protected]> Cc: <[email protected]> Subject: RTF FRM blr to pc.com cc blr-ws-134 Date: Wed, 22 Dec 2010 17:11:02 +0530 Message-ID: <[email protected]> MIME-Version: 1.0 Content-Type: multipart/mixed; boundary="----=_NextPart_000_00F7_01CBA1FB.36115580" X-Mailer: Microsoft Outlook 14.0 Content-Language: en-us X-MS-TNEF-Correlator: 00000000DCB2344DE8F50F4FBC91085BB5C06D55A4172000 thread-index: AcuhzRuTOBkvHPUnS1aLi9+cHNAWhA== Return-Path: [email protected] X-OriginalArrivalTime: 22 Dec 2010 11:41:02.0822 (UTC) FILETIME=[1C788860:01CBA1CD] This is a multipart message in MIME format. ------=_NextPart_000_00F7_01CBA1FB.36115580 Content-Type: text/plain; charset="us-ascii" Content-Transfer-Encoding: 7bit HTML Test Test Mail ------=_NextPart_000_00F7_01CBA1FB.36115580 Content-Type: application/ms-tnef; name="winmail.dat" Content-Transfer-Encoding: base64 Content-Disposition: attachment; filename="winmail.dat" // and some other code..... Any help is appreciated. Thanks.

    Read the article

  • NX Client for Windows 7 Opens Remote Desktop in Multiple Windows

    - by Corey Kennedy
    What I'm trying to do: access my Ubuntu desktop remotely via NX Client on my Windows 7 laptop. My environment: server: Ubuntu 10.10 on AMD 1Ghz/512MB RAM PC client: Windows 7 on ThinkPad sl510 Software: server is running NXServer 3.4.0. Using xfce4 window manager. Laptop is using NXClient for Windows In my NX Client "Desktop" settings I've selected "Unix" and "Custom" for OS and environment. I've also specified "startxfce4" as the application to launch when NX connects. I am able to authenticate an NX session on my laptop. By this I mean, I can start the client on my laptop, enter credentials for my Linux user, and NX establishes a connection to the server and attempts to open a remote desktop window. The problem, though, is that this remote desktop is "fragmented" into many Windows. One window will display the bulk of my desktop (complete with desktop icons for "Home," "File System," and "Trash") while another window will contain the taskbar, and another window will contain the application strip. I can select each of these Windows individually, but I cannot click on any objects within them. I've searched Super User, Ubuntu Forums, NX help, Server Fault, and tried many Google searches - none have turned up another case of this particular problem. I'm stumped. Does anyone have any suggestions for what I might try? I'm guessing the problem has to do with my xfce config files, but I've only just setup this server - it's been a long time since I've used Linux and there's a lot I just don't know. What I am NOT trying to do: use Desktop sharing from Ubuntu, whereby I VNC into a desktop that I've already established on the server. I am trying to configure this Linux box as a headless server that I can stash someplace out-of-the-way in my house, then interact with through my laptop. I don't want to have a monitor or keyboard connected to the Linux box. Thanks for your help! edit: 1/19/2011 Well, this is truly bizarre. To my knowledge I've made no changes to either system - the laptop or the server. But today after starting up the server for the first time in a few days, and making sure that nxserver was running, I was able to connect with the nxclient from my laptop with no problems. I have a full desktop in a single window and I am able to interact with it normally. This is really weird, but the problem seems to be resolved.

    Read the article

  • Can't connect to shared folders anymore?

    - by HuskyHuskie
    My home server is running Windows Server 2008 R2. I've had it running for almost a year now without any issues with shared folders. This past week I had an issue with my modem which required it to be power cycled and with that I power cycled my router. After that I haven't been able to connect to my shared network folders. I have no idea why that would even cause an issue as I've power cycled my networking equipment in the past without issues and none of my settings appear to have been lost. I am mapping these drives on my Windows 7 Ultimate machine using "Map Network Drive", from there I enter \\SERVER\Storage as I'm trying to connect to my shared folder named Storage. I receive the following error every time I try mapping the drive: Windows cannot access \\Server\Storage Check the spelling of the name. Otherwise there might be a problem with your network. To try to identify and resolve network problems, click Diagnose. Details: Error code: 0x80070035 The network path was not found. When I click Diagnose I get the following: Problems found file and print sharing resource (SERVER) is online but isn't responding to connection attempts. The remote computer isn't responding to connection on port 445, possibly due to firewall or security policy settings, or because it might be temporarily unavailable. Windows couldn't find any problems with the firewall on your computer. I've tried this from multiple computers with the same issue too. To resolve the problems so far I've tried: Disabling the firewall on SERVER Reinstalling File Services Modifying NetBT\Parameters registry values Adding a custom inbound rule for port 445 Adding port forwarding on my router for port 445 Recreating the shared folders Checking and rechecking the shared folder permissions. Resetting my user account password on the server used to access the shared folder. I'm pulling my hair out with this problem mainly because it came out of nowhere. It was working fine the night before and the next day it just stopped working. Any ideas of what I could try next are much appreciated. It should also be noted that this server is used as a web server too and that functionality still works correctly.

    Read the article

  • recommendations for efficient offsite remote backup solution of vm's

    - by senorsmile
    I am looking for recommendations for backing up my current 6 vm's(and soon to grow to up to 20). Currently I am running a two node proxmox cluster(which is a debian base using kvm for virtualization with a custom web front end to administer). I have two nearly identical boxes with amd phenom II x4's and asus motherboards. Each has 4 500 GB sata2 hdd's, 1 for the os and other data for the proxmox install, and 3 using mdadm+drbd+lvm to share the 1.5 TB's of storage between the two machines. I mount lvm images to kvm for all of the virtual machines. I currently have the ability to do live transfer from one machine to the other, typically within seconds(it takes about 2 minutes on the largest vm running win2008 with m$ sql server). I am using proxmox's built-in vzdump utility to take snapshots of the vm's and store those on an external harddrive on the network. I then have jungledisk service (using rackspace) to sync the vzdump folder for remote offsite backup. This is all fine and dandy, but it's not very scalable. For one, the backups themselves can take up to a few hours every night. With jungledisk's block level incremental transfers, the sync only transfers a small portion of the data offsite, but that still takes at least a half an hour. The much better solution would of course be something that allows me to instantly take the difference of two time points (say what was written from 6am to 7am), zip it, then send that difference file to the backup server which would instantly transfer to the remote storage on rackspace. I have looked a little into zfs and it's ability to do send/receive. That coupled with a pipe of the data in bzip or something would seem perfect. However, it seems that implementing a nexenta server with zfs would essentially require at least one or two more dedicated storage servers to serve iSCSI block volumes (via zvol's???) to the proxmox servers. I would prefer to keep the setup as minimal as possible (i.e. NOT having separate storage servers) if at all possible. I have also briefly read about zumastor. It looks like it could also do what I want, but it appears to have halted development in 2008. So, zfs, zumastor or other?

    Read the article

  • Fix/Bypass "Cannot connect to the real website-blocked" error in Google Chrome with OpenDNS blocking

    - by George H
    I have a large problem with Chrome in my organisation. I use DNS to manage web site blocking, for sites which are not appropriate and are potentially a risk to the organisation where I do this. I only want to use Chrome over the network, as Internet Explorer has compatibility problems with some sites that we use (We cannot change this either or use different sites). Therefore using internet explorer is not a solution. I do not want to install a different browser, for multiple reasons. Mainly because of the difficulty of rewriting the customised add-ons that we use. However, recently, I have had lots of problems with Chrome SSL Errors. I cannot use my custom OpenDNS block pages, which uses the contact form to request an unblocking. Chrome often blocks OpenDNS for sites (a good example is Facebook) that request HTTPS. Some sites like https://internetbadguys.com (OpenDNS example) This means that chrome refuses to load the blocking page, explaining that the site is blocked. Instead they often call IT support, but they want a solution, as they are sick of getting lots of SSL errors. I have tried looking into ways to turning this off. I have tried: Typing "proceed". That didn't work. Typing "proceed", pressing enter. Didn't work I cannot find phishing and anti-malware any more in Chrome, from the internet guides. Not using HTTPS. However there is an automatic redirect to HTTPS on most sites. Therefore the error keeps coming up. Checking my clocks. They were correct. Does anyone have an idea on how to disabling, bypassing or working around this "feature"? EDIT: This is an example what I am talking about - I found that on google images. I do not block google. EDIT 2: My clocks are correct. I cannot stop using OpenDNS either. EDIT 3: My question is: How do I stop chrome from refusing to load pages that are blocked by OpenDNS, where the server has explicitly requested HTTPS.

    Read the article

< Previous Page | 661 662 663 664 665 666 667 668 669 670 671 672  | Next Page >