Search Results

Search found 17054 results on 683 pages for 'jms request reply'.

Page 215/683 | < Previous Page | 211 212 213 214 215 216 217 218 219 220 221 222  | Next Page >

• Apache2 + FCGI + PHP5 not creating sockets / pools

  - by CodyRo

  I have a pretty vanilla setup with Apache2, FastCGI setup as DSO and serving PHP through an external CGI script that sets the max children / serves the request to PHP. The issue is FastCGI doesn't appear to be creating the PHP sockets / pooling them so each request calls the php-cgi binary, then dies off .. effectively making the reason I want to use FastCGI moot. The only configuration directives I have are: AddHandler php5-fastcgi .php Action php5-fastcgi /cgi-bin/fcgi.cgi FastCgiIpcDir /usr/local/apache2/fastcgi The dyanmic/ directory is getting created as anticipated, but there are no sockets in there. Permissions are indeed correct. Any help would be greatly appreciated, thanks!

  Read the article

• How to create an GUI that communicate with the USB Devices

  - by VINAYAK

  I am doing my Project using win 32 programming.I am just learning about win32 programming and able to create an UI.I want to communicate with an USB Device with that UI.SO,How can i go for that?Is there any predefined functions will be there are we need to create the function for communicating with the OS and get the devices List and got the details about them. My purpose is to , 1.Creating an UI that tells about the Basic information about the device(We want to send a control request to the device to get the descriptors). 2.For that first of all i want to communicate with the OS for device attachment.That will lead to get the information about the device and Enumeration takes place and then only i request the device information through descriptors by using standard Requests. 3.And also i want to create the driver for my device.That will also need to achieve for communicating with OS(Windows). So,can anyone help me about this?How can i achieve this or approach this? Note: I am at the entry level now so anyone give response will be in a detailed format like step by step process would be appreciable.

  Read the article

• Passenger, Apache and avoiding page caching

  - by user38382

  I'm hosting a rack application with passenger and apache. The application is setup to cache the content of each request to the public directory after each request. This allows apache to serve the content directly as a static page for future requests. I would like to tell Apache, presumably through some rewrite rules that any requests with query parameters should not be cached, but instead passed down to the rack application. With a mongrel setup I would just redirect it to the balancer if it meets my rewrite conditions. How do you do the same with passenger?

  Read the article

• How to log out with a command in a gnome-less environment?

  - by octosquidopus

  I installed various window managers (Awesome, dwm, etc.) from which I am not able to log out back to the login screen (gnome-session) in order to switch to another window manager. I need to reboot to do that, which is a waste of time. Question How can you log out via the terminal? didn't work.. dbus-send --session --type=method_call --print-reply --dest=org.gnome.SessionManager /org/gnome/SessionManager org.gnome.SessionManager.Logout uint32:1 ..neither did this: gnome-session-save --force-logout ..nor that: gnome-session-quit --force-logout they all returned: Failed to call logout: The name org.gnome.SessionManager was not provided by any .service files Is there a quick way to log out back into Gnome's session manager from a non-Gnome desktop manager using a terminal emulator? I know that CTRL+ALT+BACKSPACE can be configured to restart X, but I'm looking for the easiest way to log out.

  Read the article

• How to Create and Use Templates in Outlook 2010

  - by Taylor Gibb

  If you reply to the emails with the same answer over and over, it will save you a lot of time to create a template that you can use over and over. We have previously show you how to create templates in Outlook 2003, so lets take a look at using Outlook 2010. When creating a template you get started as if you were creating a new email, that is choose new email from the Home tab. You can now draft your email as normal. HTG Explains: When Do You Need to Update Your Drivers? How to Make the Kindle Fire Silk Browser *Actually* Fast! Amazon’s New Kindle Fire Tablet: the How-To Geek Review

  Read the article

• Forward Hostname via my router to another PC for debugging

  - by Markive

  Hi, My web service runs on for example: http://mydomain.com/mywebservice.asmx. This works great, but I have a PDA application which I want to debug it synchronising through this web service. Currently the only way I can do this is to debug the webservice running on the actual server which is far from ideal. What I would like to do is for any device connecting on my wireless network, if it requests mywebservice.asmx for this to forward the request to my development PC and for IIS to then handle the request and allow me to debug in Visual Studio. So device on the network that requests the hostname: mywebservice.asmx will his this PC.. I am at a loss to set this up on my router (Zoom ADSL X6), this is massively out of my scope but any help would be much appreciated

  Read the article

• No unity-greeter.conf file in /etc/lightdm/

  - by drofart

  I am running Ubuntu 12.04 with gnome-shell. I have two admin user accounts. Now, I used to know that in Precise we can set different different login wallpaper to different different user, so I tried with: sudo dbus-send --system --print-reply --dest=org.freedesktop.Accounts /org/freedesktop/Accounts/User1001 org.freedesktop.Accounts.User.SetBackgroundFile string:/usr/share/backgrounds/Green-wallpaper-27.jpg but to no avail. Then I tried sudo xhost +SI:localuser:lightdm sudo su lightdm -s /bin/bash gsettings set com.canonical.unity-greeter background '/path/to/the/wallpaper.png' It does change my login wallpaper but for all users. Then I tried to locate /etc/lightdm/unity-greeter.conf but I found that it is not there. So where is it and how can I set different login wallpapers for different users?

  Read the article

• nginx timeout albeit ridicolous configuration

  - by Joa Ebert

  The scenario is an API server that should handle uploads. Posting on my.host.com/api/upload should do something with the body the client sends. However the API server has been designed to block the whole request until it fully processed the file, including some analysis which can take up to approx. 5min (...!). This has to change of course. In the meantime I wanted to setup nginx as a load balancer in front of the API servers. I quickly ran into a timeout issue, consulted Google and came up with this ridiculous test configuration: user www-data; worker_processes 4; error_log /var/log/nginx/error.log; pid /var/run/nginx.pid; events { worker_connections 1024; } http { include /etc/nginx/mime.types; access_log off; sendfile on; send_timeout 3600; keepalive_timeout 3600 120; tcp_nopush on; tcp_nodelay on; gzip off; client_header_timeout 3600; client_body_timeout 3600; proxy_send_timeout 3600; proxy_read_timeout 3600; proxy_connect_timeout 1800; proxy_next_upstream error; include /etc/nginx/conf.d/*.conf; include /etc/nginx/sites-enabled/*; } And upstream test { server host1; server host2; } server { listen 80; server_name my.host.com; client_max_body_size 10m; location /api/ { proxy_pass http://test; proxy_set_header X-Real-IP $remote_addr; proxy_set_header X-Forwarded-For $proxy_add_x_forwarded_for; proxy_set_header Host $host; proxy_redirect off; } } Still, when an upload happens, I get the following result in the error.log: 2010/12/22 13:36:42 [error] 5256#0: *187359 upstream timed out (110: Connection timed out) while reading response header from upstream, client: xx.xx.xx.xx, server: my.host.com, request: "POST /api/upload HTTP/1.1", upstream: "http://apiserver:80/upload", host: "my.host.com" What else could I do? If I look at the log of the API server I still see that it is processing the request and analyzing the file. But I think 3600 seconds as a timeout should be more than enough. This happens even after a could of seconds. And I did a reload and force-reload of the configuration as well of course.

  Read the article

• Few New Features Added to Geekswithblogs.net

  - by Staff of Geeks

  After reviewing some of the feedback from our bloggers we added a couple new features to Geekswithblogs.net and there are still more to come.  Here is a list of the features we added.   Fixed the Twitter parser to better support URLs and Hash Tags Added some hooks behind the scenes to tags posts with common keywords automatically Added Facebook likes and Tweets to the bottom of every post Cleaned up a few skins Images on the main page for bloggers who use Gravatar or Twitter integration Random bug fixes based on Log   We are definitely working to make Geekswithblogs.net faster and better.  If you have any suggestions, please feel free to share them with the team.  On a side note, if that suggestion is move to WordPress, I will reply to you with stop writing ASP.NET for your day job and move to PHP.  That request is the equivalent in my eyes.  If we have enough bloggers leave the Microsoft .NET Platform for their main source of income, we might consider it.   Technorati Tags: Geekswithblogs.net,Features,Version 4.0

  Read the article

• Service layer coupling

  - by Justin

  I am working on writing a service layer for an order system in php. It's the typical scenario, you have an Order that can have multiple Line Items. So lets say a request is received to store a line item with pictures and comments. I might receive a json request such as { 'type': 'Bike', 'color': 'Red', 'commentIds': [3193,3194] 'attachmentIds': [123,413] } My idea was to have a Service_LineItem_Bike class that knows how to take the json data and store an entity for a bike. My question is, the Service_LineItem class now needs to fetch comments and file attachments, and store the relationships. Service_LineItem seems like it should interact with a Service_Comment and a Service_FileUpload. Should instances of these two other services be instantiated and passed to the Service_LineItem constructor,or set by getters and setters? Dependency injection seems like the right solution, allowing a service access to a 'service fetching helper' seems wrong, and this should stay at the application level. I am using Doctrine 2 as a ORM, and I can technically write a dql query inside Service_LineItem to fetch the comments and file uploads necessary for the association, but this seems like it would have a tighter coupling, rather then leaving this up to the right service object.

  Read the article

• Generating AdSense ad impression?

  - by Danny

  I owned two website. Let's say as example www.first.com and www.second.com. I have two apps in Google chrome-store whose app URL is: app1 URL = www.first.com/currency_converter.php app2 URL = www.first.com/currency_converter.php respectively. Currently if user lunch/click on first app URL then I am forwarding them to my first website homepage (www.first.com), where I have AdSense code with currency converter app. In this way I am generating Google ad impression and traffic for my first website (i.e www.first.com). So is this valid way? Does I am violating any Google AdSense rule? Please provide your reply or suggestions. P.S.: Please comment, if my question is not clear, I will try to write it in some other way.

  Read the article

• nginx + Jetty - thousands of connections stuck in LAST_ACK

  - by virulence

  I have a FreeBSD machine with jails -- two in particular, one that runs nginx and another that runs a Java program that accepts requests via Jetty (embedded mode) Jetty receives upwards of 500 requests/sec constantly and there has been an issue lately where I will constantly have over 60,000 connections in the LAST_ACK state between nginx and jetty. Distribution of all connections (includes some other services, particularly php-fpm) root@host:/root # netstat -an > conns.txt root@host:/root # cat conns.txt | awk '{print $6}' | sort | uniq -c | sort -n 18 LISTEN 112 CLOSING 485 ESTABLISHED 650 FIN_WAIT_2 1425 FIN_WAIT_1 3301 TIME_WAIT 64215 LAST_ACK Distribution of nginx - jetty connections root@host:/root # cat conns.txt | grep '10.10.1.57' | awk '{print $6}' | sort | uniq -c | sort -n 1 3 CLOSE_WAIT 3 LISTEN 18 FIN_WAIT_2 125 ESTABLISHED 64193 LAST_ACK I'd prefer every request to fully close the connection. Clients requests are about 10 minutes apart from each other so connections must be closed. Some of the connections, tcp4 0 0 10.10.1.50.46809 10.10.1.57.9050 LAST_ACK tcp4 0 0 10.10.1.50.46805 10.10.1.57.9050 LAST_ACK tcp4 0 0 10.10.1.50.46797 10.10.1.57.9050 LAST_ACK tcp4 0 0 10.10.1.50.46794 10.10.1.57.9050 LAST_ACK tcp4 0 0 10.10.1.50.46790 10.10.1.57.9050 LAST_ACK tcp4 0 0 10.10.1.50.46789 10.10.1.57.9050 LAST_ACK tcp4 0 0 10.10.1.50.46771 10.10.1.57.9050 LAST_ACK etc.. On Jetty's end I've set maxIdleTime to 2000 -- before this all connections were in ESTABLISHED but they are now LAST_ACK On Jetty's end I've set Connection: close (i.e response.setHeader(HttpHeaders.CONNECTION, HttpHeaderValues.CLOSE);) Jetty never reports a lot of open connections -- always very few. PF/IPFW is not currently being used nginx - reset_timedout_connection is on I cannot figure out how to get nginx or jetty to forcibly close the connection, is this simply something that needs to be fixed in Jetty so that it fully closes the socket after the request finishes? Thanks a lot in advance EDIT: forgot my nginx config for the proxy setup- proxy_pass http://10.10.1.57:9050; proxy_set_header HTTP_X_GEOIP $http_x_geoip; proxy_set_header GEOIP_COUNTRY_CODE $geoip_country_code; proxy_set_header X-Real-IP $remote_addr; proxy_set_header X-Forwarded-For $proxy_add_x_forwarded_for; proxy_set_header Host $http_host; proxy_set_header Connection ""; proxy_http_version 1.1; EDIT2: Forcing Jetty to close the connection via request.getConnection().getEndPoint().close() does nothing -- it's obvious the connection IS being closed (as it's in LAST_ACK) but why isn't it getting past this? Is Nginx keeping the connection open to the backend for some reason?

  Read the article

• Nginx server_name is set to mydomain.com, so why is www.mydomain.com getting served too?

  - by Lorenz Forvang

  I have my Nginx conf set up as follows: server { listen 443 ssl; server_name mydomain.com; ... } When I load https://mydomain.com, the site loads fine. But when I load https://www.mydomain.com, the site loads as well. Why is this happening? I set up the DNS records using Amazon Route 53 as: A mydomain.com xxx.xxx.xxx.xxx (IP) CNAME www.mydomain.com mydomain.com So is a request to www.mydomain.com arriving at Nginx as a request to mydomain.com? If so, how do I differentiate requests to www.mydomain.com and mydomain.com at my server?

  Read the article

• Hyperic HQ metrics not working

  - by Robin Weston

  I am having a problem with Hyperic HQ. Several metrics, some in IIS 6.x (Request Execution Time, Request Wait Time) and some in .NET 2.0 (Bytes in all Heaps, Exceptions Thrown per Minute), always show 0. If I view perfmon on the server itself I can see that the counters have values greater than zero. There are some metrics that work fine, such as Total Get Requests per Minute and the other IIS defaults. I have looked in the server logs but nothing obvious shows up. Please advise. Am happy to find more information if required.

  Read the article

• Is realtime validation of username good or bad?

  - by iamserious

  I have a simple form for the user to sign up to my site; with email, username and password fields. We are now trying to implement an ajax validation so the user doesn't have to post the form to find out if the username is already taken. I can do this either on keyup event or on text blur event. My question is, which of these is really the best way to do? Keyup From the user POV, it would be good if the validation is done as and when they are typing, (on key up event) - of course, I am waiting for half a second to see if the user stops typing before firing off the request, and user can make any adjustments immediately. But this means I am sending way more requests than if I validated the username on Blur event. Blur The number of requests will be much lower when the validation is done on blur event, But this means the user has to actually go away from the textbox, look at the validation result, and if necessary go back to it to make any changes and repeat the whole process until he gets it right. I had a quick look at google, tumblr, twitter and no one actually does username validations on keyup events, (heck, tubmlr waits for the form to be posted) but I can swear I have seen keyup validations in a lot of places too. So, coming back to the question, will keyup validations be too many for server, is it an unnecessary overhead? or is it worth taking these hits to give user a better experience? ps: all my regex validations etc are already done on javascript and only when it passes all these other criteria does it send a request to server to check if a username already exists. (And the server is doing a select count(1) from user where username = '' - nothing substantial, but still enough to occupy some resource) pps: I'm on asp.net, MS SQL stack., if that matters.

  Read the article

• thick client migration to web based application

  - by user1151597

  This query is related to application design the technology that I should consider during migration. The Scenario: I have a C#.net Winform application which communicates with a device. One of the main feature of this application is monitoring cyclic data(rate 200ms) sent from the device to the application. The request to start the cyclic data is sent only once in the beginning and then the application starts receiving data from the device until it sends a stop request. Now this same application needs to be deployed over the web in a intranet. The application is composed of a business logic layer and a communication layer which communicates with the device through UDP ports. I am trying to look at a solution which will allow me to have a single instance of the application on the server so that the device thinks that it is connected as usual and then from the business logic layer I can manage the clients. I want to reuse the code of the business layer and the communication layer as much as possible. Please let me know if webserives/WCF/ etc what i should consider to design the migration. Thanks in advance.

  Read the article

• Windows Advanced Firewall certificate based IPSEC

  - by Tim Brigham

  I'm working on migrating from using IPSEC settings stored under the 'IP Security Policies on Active Directory' to using the 'Windows Firewall with Advanced Security' for my 2008+ boxes. I have successfully been able to get this set up using Kerberos authentication, however my openswan implementation on my Linux boxes is using certificates. Whenever I try changing the authentication method to computer certificate (using RSA and my root CA) the connection is bombing out. I've made this change at both a connection request policy and on the IPSEC settings on the root Windows Firewall with Advanced Security node. The windows event log shows the authentication request is taking place but failing negotiating a mode. What am I missing here?

  Read the article

• chmod 700 and htaccess deny from all enough?

  - by John Jenkins

  I would like to protect a public directory from public view. None of the files will ever be viewed online. I chmoded the directory to 700 and created an htaccess file that has "deny from all" inside it. Is this enough security or can a hacker still gain access to the files? I know some people will say that hackers can get into anything, but I just want to make sure that there isn't anything else I can do to make it harder to hack. Reply: I am asking if chmod 700 and deny from all is enough security alone to prevent hackers from getting my files. Thanks.

  Read the article

• Spreading incoming batched data into a real-time stream

  - by pr1001

  I would like to display some events in 'real-time'. However, I must fetch the data from another source. I can request the last X minutes, though the source is updated approximately every 5 minutes. This means that there will be a delay between the most recent data retrieved and the point in time that I make the request. Second, because I will be receiving a batch of data, I don't want to just fire out all the events down a socket once my fetcher has retrieved it: I would like to spread out the events so that they are both accurately spaced amongst each other and in sync with their original occurrences (e.g. an event is always displayed 6 minutes after it actually happened). My thought is to fetch the data every 5 minutes from the source, knowing that I won't get the very latest data. The original data would be then queued to be sent down the socket 7.5 minutes from its original timestamp – that is, at least ~2.5 minutes from when its batch was fetched and at most 7.5 minutes since then. My question is this: is this the best way to approach the problem? Does this problem have any standard approaches or associated literature related to implementation best-practices and edge cases? I am a bit worried that the frequency of my fetches and the frequency in which the source is updated will get out of sync, leading to points where no data will be retrieved from the source. However, since my socket delay is greater than my fetch frequency, the subsequent fetch should retrieve newer data before the socket queue is empty. Is that correct? Am I missing something? Thanks!

  Read the article

• Ubuntu getting wrong hostname from DHCP

  - by sam

  When provisioning new Ubuntu Precise (12.04) servers, the hostname they're getting seems to be generated from the DNS search path, not a reverse lookup on the hostname. Take the following configuration BIND is configured with the hostname, and reverse name Normal zone $TTL 600 $ORIGIN srv.local.net. @ IN SOA ns0.local.net. hostmaster.local.net. ( 2014082101 10800 3600 604800 600 ) @ IN NS ns0.local.net. @ IN MX 5 mail.local.net. my-new-server IN A 10.32.2.30 And reverse @ IN SOA ns0.local.net. hostmaster.local.net. ( 2014082101 10800 3600 604800 600 ) @ IN NS ns0.local.net. $ORIGIN 32.10.in-addr.arpa. 30.2 IN PTR my-new-server.srv.local.net. Then DHCPD is configured to hand out static leases based on mac addresses like so subnet 10.32.2.0 netmask 255.255.254.0 { option subnet-mask 255.255.254.0; option routers 10.32.2.1; option domain-name-servers 10.32.2.1; option domain-name "util.of1.local.net of1.local.net srv.local.net"; site-option-space "pxelinux"; option pxelinux.magic f1:00:74:7e; if exists dhcp-parameter-request-list { option dhcp-parameter-request-list = concat(option dhcp-parameter-request-list,d0,d1,d2,d3); } group { option pxelinux.configfile "pxelinux.cfg/pxeboot"; host my-new-server { fixed-address my-new-server.srv.local.net; hardware ethernet aa:aa:aa:bb:bb:bb; } } } So the hostname should be my-new-server.srv.local.net, however when building a Ubuntu 12.04 node, the hostname ends up as my-new-server.util.of1.local.net When building Lucid (10.04) hosts, the hostname will be correct, it's only on Precise/12.04 nodes we have the problem. Doing a normal and reverse lookup on the host and IP returns the correct result Sams-MacBook-Pro:~ sam$ host my-new-server my-new-server.srv.local.net has address 10.32.2.30 Sams-MacBook-Pro:~ sam$ host my-new-server.srv.local.net my-new-server.srv.local.net has address 10.32.2.30 Sams-MacBook-Pro:~ sam$ host 10.32.2.30 30.2.32.10.in-addr.arpa domain name pointer my-new-server.srv.local.net. The contents of the hosts file is incorrect too 127.0.0.1 localhost 127.0.1.1 my-new-server.util.of1.local.net of1.local.net srv.local.net my-new-server So it looks like when it creates the hosts file, it puts the entire contents of the DNS search path into the local address so the FQDN according to the server is the short hostname as defined, then the first domain in the search path. Is there a way to get around this behaviour, or fix this so it gets the hostname correctly? It's picking up the first part of the hostname, then the rest is wrong.

  Read the article

• HTTP Caching Server that supports POST

  - by Jeroen

  I am hosting a REST service which is sending appropriate cache-control headers. I use Varnish as a caching server in front of my webserver. However, a limitation of varnish is that it doesn't support caching HTTP POST and HTTP PUT. Is there any alternate caching server that will be able to cache these requests? I understand that caching POST is a bit tricky because you cannot just cache based on the url as a key like for GET; it needs to actually inspect the request body. In case of multipart/form-data requests, there should probably be a limit on the size of the request body for it to be cached (so that big file uploads, etc won't be cached). Nevertheless I really want to be able to cache short HTTP POST, or at least the application/x-www-form-urlencoded ones.

  Read the article

• Bridging two sockets

  - by Itehnological

  I wondered if it is possible to bridge two incoming tcp sockets. For example: Client A -----> Server <----- Client B The the server sends it's magic to both clients and then they connect to each other bypassing the server Server Client A ----------><---------- Client B UPDATE: The idea is when those clients can't bind to ports to listen to still be able to create connection between each other with the help of the server. For example Client A and Client B have tcp sockets with the server. User A decides to chat with User B and creates a new tcp connection with the server with the request to bridge it with User B. The server sends that request to Client B and it also opens up a new tcp connection with the server for that chat line. Now when the server has both chat connections from A and B it bridges them and they can work without the server, and as a result the server won't have to process all the messages and files the two users share. That's the idea/

  Read the article

• Generate a Word document from list data

  - by PeterBrunone

  This came up on a discussion list lately, so I threw together some code to meet the need.  In short, a colleague needed to take the results of an InfoPath form survey and give them to the user in Word format.  The form data was already in a list item, so it was a simple matter of using the SharePoint API to get the list item, formatting the data appropriately, and using response headers to make the client machine treat the response as MS Word content.  The following rudimentary code can be run in an ASPX (or an assembly) in the 12 hive.  When you link to the page, send the list name and item ID in the querystring and use them to grab the appropriate data. // Clear the current response headers and set them up to look like a word doc.HttpContext.Current.Response.Clear();HttpContext.Current.Response.Charset ="";HttpContext.Current.Response.ContentType ="application/msword";string strFileName = "ThatWordFileYouWanted"+ ".doc";HttpContext.Current.Response.AddHeader("Content-Disposition", "inline;filename=" + strFileName);// Using the current site, get the List by name and then the Item by ID (from the URL).string myListName = HttpContext.Current.Request.Querystring["listName"];int myID = Convert.ToInt32(HttpContext.Current.Request.Querystring["itemID"]);SPSite oSite = SPContext.Current.Site;SPWeb oWeb = oSite.OpenWeb();SPList oList = oWeb.Lists["MyListName"];SPListItem oListItem = oList.Items.GetItemById(myID);// Build a string with the data -- format it with HTML if you like. StringBuilder strHTMLContent = newStringBuilder();// *// Here's where you pull individual fields out of the list item.// *// Once everything is ready, spit it out to the client machine.HttpContext.Current.Response.Write(strHTMLContent);HttpContext.Current.Response.End();HttpContext.Current.Response.Flush();

  Read the article

• How to collaborate on features using github

  - by Robert Dailey

  github encourages 1 fork per user, so that that user can work independently on a feature and then request that feature to be accepted into the main repository via pull request. However, what if 2 developers need to collaborate on that feature? What is the ideal workflow for this? I could see a number of options: Both developers fork the original repository. Each developer pulls/pushes changes between each other's repository. This seems like a lot of work (tiny micro operations) and also creates a delay between changes, so increases the window for conflicts. Developer 1 forks from the main repository, developer 2 forks from developer 1. Same as #1 mainly but hopefully simplifies Developer 2's life a little? Developer 1 gives Developer 2 permissions to his own fork, so they both work out of the same central repository. Not sure if this is ideal. I'm also curious where branches come into this. Obviously there would be a branch for the feature itself but that branch can't exist in a single place, it would have to exist on multiple forks and be synchronized. Basically just really confused about this workflow, would like an approach for how this can be best accomplished.

  Read the article

• Why is domU faster than dom0 on IO?

  - by Paco

  I have installed debian 7 on a physical machine. This is the configuration of the machine: 3 hard drives using RAID 5 Strip element size: 1M Read policy: Adaptive read ahead Write policy: Write Through /boot 200 MB ext2 / 15 GB ext3 SWAP 10GB LVM rest (~500GB) emphasized text I installed postgresql, created a big database (over 1GB). I have an SQL request that takes a lot of time to run (a SELECT statement, so it only reads data from the database). This request takes approximately 5.5 seconds to run. Then, I installed XEN, created a domU, with another debian distro. On this OS, I also installed postgresql, with the same database. The same SQL request takes only 2.5 seconds to run. I checked the kernel on both dom0 and domU. uname-a returns "Linux debian 3.2.0-4-amd64 #1 SMP Debian 3.2.41-2+deb7u2 x86_64 GNU/Linux" on both systems. I checked the kernel parameters, which are approximately the same. For those that are relevant, I changed their values to make them match on both systems using sysctl. I saw no changes (the requests still take the same amount of time). After this, I checked the file systems. I used ext3 on domU. Still no changes. I installed hdparm, and ran hdparm -Tt on both systems, on all my partitions on both systems, and I get similar results. Now, I am stuck, I don't know what is different, and what could be the cause of such a big difference. Additional Info: Debian runs on a Dell server PowerEdge 2950 postgresql: 9.1.9 (both dom0 and domU) xen-linux-system: 3.2.0 xen-hypervisor: 4.1 Thanks EDIT: As Krzysztof Ksiezyk suggested, it might be due to some file caching system. I ran the dd command to test both the read and write speed. Here is domU: root@test1:~# dd if=/dev/zero of=/root/dd count=5MB bs=1MB ^C2020+0 records in 2020+0 records out 2020000000 bytes (2.0 GB) copied, 18.8289 s, 107 MB/s root@test1:~# dd if=/root/dd of=/dev/null count=5MB bs=1MB 2020+0 records in 2020+0 records out 2020000000 bytes (2.0 GB) copied, 15.0549 s, 134 MB/s And here is dom0: root@debian:~# dd if=/dev/zero of=/root/dd count=5MB bs=1MB ^C1693+0 records in 1693+0 records out 1693000000 bytes (1.7 GB) copied, 8.87281 s, 191 MB/s root@debian:~# dd if=/root/dd of=/dev/null count=5MB bs=1MB 1693+0 records in 1693+0 records out 1693000000 bytes (1.7 GB) copied, 0.501509 s, 3.4 GB/s What can be the cause of this caching system? And how can we "fix" it? Can we apply it to dom0? EDIT 2: I switched my virtual disk type. To do so I followed this article. I did a dd if=/dev/vg0/test1-disk of=/mnt/test1-disk.img bs=16M Then in /etc/xen/test1.cfg, I changed the disk parameter to use file: instead of phy: it should have removed the file caching, but I still get the same numbers (domU being much faster for Postgres)

  Read the article

< Previous Page | 211 212 213 214 215 216 217 218 219 220 221 222  | Next Page >