Search Results

Search found 10755 results on 431 pages for 'cluster shared volume'.

Page 37/431 | < Previous Page | 33 34 35 36 37 38 39 40 41 42 43 44  | Next Page >

• Apache2 server does not start cannot pen shared object file

  - by sid__

  I am working with Apache and Passenger for a Rails project. And a during a restart I got the following error Cannot load /usr/lib/ruby/gems/1.8/gems/passenger-2.2.11/ext/apache2/mod_passenger.so into server: /usr/lib/ruby/gems/1.8/gems/passenger-2.2.11/ext/apache2/mod_passenger.so: cannot open shared object file: No such file or directory However there is no change in the apache configuration file. I have attached the snippet from the conf file 287 LoadModule passenger_module /usr/lib/ruby/gems/1.8/gems/passenger-2.2.11/ext/apache2/mod_passenger.so 288 PassengerRoot /usr/lib/ruby/gems/1.8/gems/passenger-2.2.11 289 PassengerRuby /usr/bin/ruby1.8 I am also unable to locate the shared object file in the location pointed to by the server though I am not really sure how the .so file is works (created/destroyed) I would also appreciate it if someone could explain to me what exactly has happened. I understand the shared object file is mission, what could be the reason it got deleted.

  Read the article

• Wrong owner and group for files created under a samba shared directory

  - by agmao

  I am trying to make writing to a shared samba directory work. I got a very weird problem. Now the shared directory is writable from a client machine. But the files created under the samba share directory have weird owner and group names. I am writing to the shared directory as user mike under the client machine, but the file created always has user and group name as steve instead... Does anybody know why that would happen...? Another thing I just noticed is that on the samba server, the files have owner and user name as samba, which I created for samba clients. Thanks a lot

  Read the article

• VPN pre-shared key problems

  - by Owl

  I have two vpns set up on a Symantec Gateway Security 320. VPN 1 goes to a Symantec Firewall/VPN 100 to another clinic of ours and every hour they lose connectivity and the error log on the Firewall/VPN100 shows an invalid pre-shared key error, although, both devices show the same pre-shared key entered. VPN 2 goes to our software vendor to use an additional part of our program. I am unable to ping the remote address and so is the other company, but my VPN status shows it is connected. They have told me the pre-shared key seemed to be automatically trying to resubmit itself as if it were incorrect, about every hour even though it is correct. They also told me port80 traffic was closed but I show the HTTP service using 80 redirected to 80 in my firewall settings. Please help.

  Read the article

• VirtualBox Shared Folder encoding issue

  - by Somebody

  I'm using Ubuntu in Virtualbox and have a shared folder mounted to Virtualbox which i'm accessing inside Ubuntu. The problem is, that when i'm editing and saving some files from shared folder in Windows it's getting some strange symbols at the end of edited file. There must be some encoding issues. Doesn't Virtualbox automatically converts files to Unix standards? To fix that, i have to re-mount shared folder inside Ubuntu each time i'm editing some file. Any solution to avoid re-mounting each time I edit? I'm mounting like that: mount -t vboxsf SVN /opt/htdocs/ Thanks.

  Read the article

• Windows 7: Shared folder over wifi working for ONLY "Guests"

  - by James_Smith

  Hi, I have a desktop and a laptop connected to same wifi router. Desktop is connected with wire and laptop with wifi. Both the system runs windows 7 and are on the same workgroup. I have shared some folders on desktop and can view the shared folder list on laptop under "network places". But when I try to open a folder, a prompt appears for username and password. When I enter BOTH username and password, It does not authenticate when I enter ONLY username, I get:- "Windows cannot access \WIn71\Setups you do not have permission to access ..." Now to get around this message I have to give access to "Guests" group on my desktop shared folder. Cant seem to figure out why It cant authenticate username with password. And giving access to "Guests" does not sounds safe!

  Read the article

• Bridging Virtual Networking into Real LAN on a OpenNebula Cluster

  - by user101012

  I'm running Open Nebula with 1 Cluster Controller and 3 Nodes. I registered the nodes at the front-end controller and I can start an Ubuntu virtual machine on one of the nodes. However from my network I cannot ping the virtual machine. I am not quite sure if I have set up the virtual machine correctly. The Nodes all have a br0 interfaces which is bridged with eth0. The IP Address is in the 192.168.1.x range. The Template file I used for the vmnet is: NAME = "VM LAN" TYPE = RANGED BRIDGE = br0 # Replace br0 with the bridge interface from the cluster nodes NETWORK_ADDRESS = 192.168.1.128 # Replace with corresponding IP address NETWORK_SIZE = 126 NETMASK = 255.255.255.0 GATEWAY = 192.168.1.1 NS = 192.168.1.1 However, I cannot reach any of the virtual machines even though sunstone says that the virtual machine is running and onevm list also states that the vm is running. It might be helpful to know that we are using KVM as a hypervisor and I am not quite sure if the virbr0 interface which was automatically created when installing KVM might be a problem.

  Read the article

• GlassFish cluster-targeted jdbc is not enabled

  - by Jin Kwon

  I have a GlassFish cluster. When I tried to add node and a instance, DAS saids a bunch of error messages telling Resource [ jdbc/xxxx ] of type [ jdbc ] is not enabled [#|2012-11-14T12:07:04.318+0900|SEVERE|glassfish3.1.2|javax.enterprise.system.core.com.sun.enterprise.v3.server|_ThreadID=2803;_ThreadName=Thread-2;|java.lang.StackOverflowError at java.io.FileOutputStream.writeBytes(Native Method) at java.io.FileOutputStream.write(FileOutputStream.java:318) at java.io.BufferedOutputStream.write(BufferedOutputStream.java:122) at java.io.PrintStream.write(PrintStream.java:480) at sun.nio.cs.StreamEncoder.writeBytes(StreamEncoder.java:221) at sun.nio.cs.StreamEncoder.implFlushBuffer(StreamEncoder.java:291) at sun.nio.cs.StreamEncoder.implFlush(StreamEncoder.java:295) at sun.nio.cs.StreamEncoder.flush(StreamEncoder.java:141) at java.io.OutputStreamWriter.flush(OutputStreamWriter.java:229) at java.util.logging.StreamHandler.flush(StreamHandler.java:242) at java.util.logging.ConsoleHandler.publish(ConsoleHandler.java:106) at java.util.logging.Logger.log(Logger.java:522) at com.sun.logging.LogDomains$1.log(LogDomains.java:372) at java.util.logging.Logger.doLog(Logger.java:543) at java.util.logging.Logger.log(Logger.java:607) at com.sun.enterprise.resource.deployer.JdbcResourceDeployer.deployResource(JdbcResourceDeployer.java:117) at org.glassfish.javaee.services.ResourceProxy.create(ResourceProxy.java:90) at com.sun.enterprise.naming.impl.SerialContext.lookup(SerialContext.java:507) at com.sun.enterprise.naming.impl.SerialContext.lookup(SerialContext.java:455) at javax.naming.InitialContext.lookup(InitialContext.java:411) at javax.naming.InitialContext.lookup(InitialContext.java:411) at com.sun.appserv.connectors.internal.api.ResourceNamingService.lookup(ResourceNamingService.java:221) the JDBC Resource is ok and targeted with the cluster. I've installed the JDBC driver on the new node. Can anybody help?

  Read the article

• How To Set Up A Loadbalanced High-Availability Apache Cluster On Windows

  - by bReAd

  Setting up a two-node Apache web server cluster that provides high-availability. In front of the Apache cluster we create a load balancer that splits up incoming requests between the two Apache nodes. Because we do not want the load balancer to become another “Single Point Of Failure”, we must provide high-availability for the load balancer, too. Therefore our load balancer will in fact consist out of two load balancer nodes that monitor each other using heartbeat, and if one load balancer fails, the other takes over silently. The following setup is proposed: Apache node 1: webserver1.example.com (webserver1) – IP address: 192.168.0.101; Apache document root: /var/www Apache node 2: webserver2.example.com (webserver2) – IP address: 192.168.0.102; Apache document root: /var/www Load Balancer node 1: loadb1.example.com (loadb1) – IP address: 192.168.0.103 Load Balancer node 2: loadb2.example.com (loadb2) – IP address: 192.168.0.104 Virtual IP Address: 192.168.0.105 (used for incoming requests) Currently, there are many solutions for Linux machines and there aren't any on windows. I've tried searching a long time for solutions on Windows platform How do I create the virtual IP in windows and perform monitoring and make the load balancer listen to the virtual IP Address?

  Read the article

• Clustering/load balancing for cluster unaware applications

  - by AaronLS

  Forgive me if I use any of these terms incorrectly. I am wondering if there is any kind of software that would allow my two "join" two computers together such that a cluster unaware application could utilize their combined computing resources? By "cluster unaware" I mean an application that isn't designed to share work across multiple services. My understanding is that clustering is enabled by the specific application by it's architecture, such that messaging with multiple instances of the application coordinate the sharing of work. Instead I am looking for something that enables clustering at the OS or virtualization level, so that any application could essentially be clustered. Failing that, I am also wondering about the following scenario: We have 3 different applications we will call A, B, and C. We have 2 single core computers. At any given time lets say that any combination of those applications will be CPU intensive. In cases where only 2 of those apps are very active, have one of them moved over to a different server. In a nutshell, some sort of dynamic automatic shuffling of the application's load. I have heard of virtual machines that can be migrated across physical machines while live, but I am wondering if this can be done automatically in response to an application's or VM's CPU activity?

  Read the article

• Linux HA / cluster: what are the differences between Pacemaker, Heartbeat, Corosync, wackamole?

  - by Continuation

  Can you help me understand Linux HA? Pacemaker, Heartbeat, Corosync seem to be part of a whole HA stack, but how do they fit together? How does wackamole differ from Pacemaker/Heartbeat/Corosync? I've seen opinions that wackamole is better than Heartbeat because it's peer-based. Is that valid? The last release of wackamole was 2.5 years ago. Is it still being maintained or active? What would you recommend for HA setup for web/application/database servers?

  Read the article

• PostgreSQL failover cluster on Windows Server

  - by user36997

  We are looking for advice on how to setup a basic failover cluster for our application: We will be using 4 machines running Microsoft Windows Server (most probably 2003). All four will always run our application, which is essentially a web service. Load balancing is "outsourced" - somebody else handles the distribution of the web requests among the servers. Only one of the servers will be running the PostgreSQL server actively at any given time. Another server (of the four) also has the DB installed, but is on standby/passive. The DB data is stored on shared storage. No copying data between servers. Reads are done very frequently by many end-users, and in rather small chunks of data. Writes are done much less frequently, by less users, and in very large bulks of data. Now, how can one configure Microsoft Cluster Service to keep only one instance of the DB server and 4 instances (1 per server) of our application at all times? And does PostgreSQL integrate neatly with MSCS at all? Update: Instead of keeping the data on shared storage, I also consider using log shipping to replicate data on a couple of DB servers. There are two issues with this option: Log shipping only makes sure that I have a second server that gets all of the data and is ready to take over. How do I implement the actual failure detection and failover switch? Switching back: Suppose the master fails and the system automatically fails over to the slave, and later the master comes back online. I understand that with WAL shipping this will require to reconfigure the log shipping once again, and that switching back is far from seamless. Is that so?

  Read the article

• RAID Volume is no longer showing in Raid Controller BIOS and in Windows

  - by Gordon

  Hi all, I have installed some critical Windows Updates yesterday and now my external RAID Volume no longer shows in Windows Vista x64. All updates went through successfully. For their description, I cannot see how they should relate to the issue, but this is the only change that happened, so who knows. Anyway, here is the details: I have an external eSata enclosure that is running on a SiI4726 controller. I can connect to the controller with it's management utility from the computer the enclosure is connected to. The three drives in the enclosure show up as JBODs. I had those drives configured to be one logical RAID5 drive. RAID management is done through a SiI3132 SoftRaid controller. The Raid Management Utility just shows empty channels where it usually shows the Raid Group. In the Windows Disk Manager, I can see an unknown unitialized device. This is fine according to the setup manual. What it doesn't show is my Raid drive. It's gone. Also, when booting Windows, the BIOS of the controller used to show the RAID volume before booting the OS. This is not happening anymore. Updating drivers and firmware did not help. I have made sure the drivers and firmware are compatible to each others. And like I said, it used to work before. Any clues?

  Read the article

• Virtual machines interconnection inside Proxmox 2.1 Cluster

  - by Anton

  We have 3 physical servers (each with 1 NIC) in different datacentres, all of them are interconnected by openvpn bridged private network (10.x.x.x). Inside this network we have fully functional 3 nodes Proxmox 2.1 cluster. So, actually question is: Is there any "proper" way to make "global" local network (172.16.x.x) for all VMs inside cluster, so even if we move VM from one node to other we could reach it by static IP regardless of it's physical location? BTW, we can't add dedicated NIC to each server. Thanks in advance. EDIT: I have tried to make a separate openvpn bridge for 172.16.x.x, now I have at each server two interfaces: SRV1: openvpnbr1 - 172.16.13.1 vmbr0 - 172.16.1.1 SRV2: openvpnbr1 - 172.16.13.2 vmbr0 - 172.16.2.1 But now there is no connection between those ifaces: SRV1: ping 172.16.13.2 From 172.16.1.1 icmp_seq=2 Destination Host Unreachable SRV2: ping 172.16.13.1 From 172.16.2.1 icmp_seq=2 Destination Host Unreachable If I shut down vmbr0 interfaces, so there is connection between servers over openvpn, but vmbr0 is used by Proxmox... Where I am wrong?

  Read the article

• SBD killing both cluster nodes when there are even small SAN network problems

  - by Wieslaw Herr

  I am having problems with stonith SBD in a openais-based cluster. Some background: The active/passive cluster has two nodes, node1 and node2. They are configured to provide an NFS service to users. To avoid problems with split-brain, they are both configured to use SBD. SBD is using two 1MB disks available to the hosts via an multipath fibre-channel network. The problems start if something happens with the SAN network. For example, today one of the brocade switches got rebooted and both nodes lost 2 out of 4 paths to each disks, which resulted in both nodes committing suicide and rebooting. This, of course, was highly undesirable because a) there were paths left b) even if the switch would be out for 10-20 seconds a reboot cycle of both nodes would take 5-10 minutes and all NFS-locks would be lost. I tried increasing the SBD timeout values (to 10sec+ values, dump attached at the end), however a "WARN: Latency: No liveness for 4 s exceeds threshold of 3 s" hints that something isn't working as I would it expect to. Here is what I would like to know: a) Is SBD working as it should killing nodes when 2 paths are available? b) If not, is the multipath.conf file attached correct? The storage controller we use is an IBM SVC (IBM 2145), should there be any specific configuration for it? (as in multipath.conf.defaults) c) How should I go about increasing the timeouts in SBD attachements: Multipath.conf and sbd dump (http://hpaste.org/69537)

  Read the article

• LVM2 volume group lost

  - by MrG

  I updated one of my servers, but - although I took care not to modify - the volume groups on /dev/sdb1 were lost, although the physical volumes seem to be still there: [root@server ~]# pvscan PV /dev/sda2 VG VolGroup lvm2 [465,16 GiB / 0 free] PV /dev/sdb1 lvm2 [1,82 TiB] Total: 2 [2,27 TiB] / in use: 1 [465,16 GiB] / in no VG: 1 [1,82 TiB] [root@server ~]# pvs -v Scanning for physical volume names PV VG Fmt Attr PSize PFree DevSize PV UUID /dev/sda2 VolGroup lvm2 a-- 465,16g 0 465,16g HftbaD-MBs0-3p7D-6O13-CrzU-T9Gb-6W0ofB /dev/sdb1 lvm2 a-- 1,82t 1,82t 1,82t dD4XZP-WStA-61xV-5Sff-ifmW-R4rR-JenHoU [root@server ~]# pvck -d -v /dev/sdb1 Scanning /dev/sdb1 Found label on /dev/sdb1, sector 1, type=LVM2 001 Found text metadata area: offset=4096, size=1044480 Found LVM2 metadata record at offset=10752, size=1037824, offset2=0 size2=0 Found LVM2 metadata record at offset=9216, size=1536, offset2=0 size2=0 Found LVM2 metadata record at offset=7168, size=2048, offset2=0 size2=0 Found LVM2 metadata record at offset=5632, size=1536, offset2=0 size2=0 I attempted to fix it as described here and was able to extract the 4 meta data sets listed above (using i.e. dd bs=1 skip=5632 count=1536 if=/dev/sdb1 of=output.file), none of them includes the lv_data which I'm missing. Please advise how I could access the files which should be on /dev/sdb1 there. Any help is appreciated!

  Read the article

• Join multiple filesystems (on multiple computers) into one big volume

  - by jm666

  Scenario: Have 10 computers, each have 12x2TB HDDs (currently) in raidZ2 (10+2) configuration, so, in the each computer i have one approx. 20TB volume. Now, need those 10 separate computers (separate raid groups) join into one big volume. What is the recommended solution? I'm thinking about the FCoE (10GB ethernet). So, buying into each computer FCoE (10GB ethernet card) and - what need more on the hardware side? (probably another computer, FCoE switch? like Cisco Nexus?) The main question is: what need to install and configure on each computer? Currently they have freebsd/raidz2, but it is possible change it into Linux/Solaris if needed. Any helpful resource what talking about how to build a big volumes from smaller raid-groups (on the software side) is very welcomed. So, what OS, what filesystem, what software - etc. In short: want get one approx. 200TB storage (in one filesystem) from already existing computers/storage. Don't need fast writes, but need good performance on reading data. (as a big fileserver), what will works transparently, so when storing data don't want care about onto what computer the data goes. (e.g. not 10 mountpoints - but one big logical filesystem). Thanks.

  Read the article

• using i7 "gamer" cpu in a HPC cluster

  - by user1219721

  I'm running WRF weather model. That's a ram intensive, highly parallel application. I need to build a HPC cluster for that. I use 10GB infiniband interconnect. WRF doesn't depends of core count, but on memory bandwidth. That's why a core i7 3820 or 3930K performs better than high-grade xeons E5-2600 or E7 Seems like universities uses xeon E5-2670 for WRF. It costs about $1500. Spec2006 fp_rates WRF bench shows $580 i7 3930K performs the same with 1600MHz RAM. What's interesting is that i7 can handle up to 2400MHz ram, doing a great performance increase for WRF. Then it really outperforms the xeon. Power comsumption is a bit higher, but still less than 20€ a year. Even including additional part I'll need (PSU, infiniband, case), the i7 way is still 700 €/cpu cheaper than Xeon. So, is it ok to use "gamer" hardware in a HPC cluster ? or should I do it pro with xeon ? (This is not a critical application. I can handle downtime. I think I don't need ECC?)

  Read the article

• Volume is no longer showing in Raid Controller BIOS and in Windows

  - by Gordon

  Hi all, I have installed some critical Windows Updates yesterday and now my external RAID Volume no longer shows in Windows Vista x64. All updates went through successfully. From their description, I cannot see how they should relate to the issue, but this is the only change that happened, so who knows. Anyway, here is the details: I have an external eSata enclosure that is running on a SiI4726 controller. I can connect to the controller with it's management utility from the computer the enclosure is connected to. The three drives in the enclosure show up as JBODs. I had those drives configured to be one logical RAID5 drive. RAID management is done through a SiI3132 SoftRaid controller. The Raid Management Utility just shows empty channels where it usually shows the Raid Group. In the Windows Disk Manager, I can see an unknown unitialized device. This is fine according to the setup manual. What it doesn't show is my Raid drive. It's gone. Also, when booting Windows, the BIOS of the controller used to show the RAID volume before booting the OS. This is not happening anymore. Updating drivers and firmware did not help. I have made sure the drivers and firmware are compatible to each others. And like I said, it used to work before. Any clues?

  Read the article

• Setting up a global MySQL Cluster in the cloud

  - by GregB

  I'm giving the question an overhaul to more specifically identify where I need help. I use two tools to manage a bunch of cloud server: Puppet and Rundeck. Both of these can be configured to use a mysql backend. I'd like to setup an instance of each application in both the U.S., and the U.K., treating the U.K. servers as hot stand-bys in case of failure in the U.S. I want to use a MySql cluster so that the data is automatically replicated from the U.S. to the U.K. Because these are hot standbys, high performance is not a goal. Redundancy and data integrity are most important. My question revolves around the setup of the mysql cluster. I want to run three servers, each one running a data node, a sql node, and a management node. Is this a valid configuration for mysql server? If so, could someone point me in the right direction for creating such a setup? I've downloaded the offical tarball, and the official debian, and the documentation for them contradicts many of the online tutorials. I'm installing on Ubuntu 10.04.

  Read the article

• How to boost playback volume in real time on media recorded with a very low volume.

  - by L Marksman

  I have never heard a satisfactory answer to this often misunderstood question, let me explain. Lets say I have a sound card and earphones/speakers that can play back audio loud enough in most cases. This is great but the problem is that you always find people who do not know how to record audio, from Youtube video's to music. So now you end up with a audio playback that only uses 10% or less of the capacity of your sound hardware, in vista/win 7 you will see this frequently in the mixer with the volume pushed up to max but the green sound level only goes up a millimeter or two. I am looking for (preferably free) software or a method to boost the sound level of any audio from any source in real time to use more of my hardware capacity similar to what VLC media player can do. Oh and please, do not tell me it is impossible. I am not trying to boost the volume past what my hardware is capable of, I am just trying to use my hardware's full capacity. Also please do not tell met to buy new hardware, I know I can use hardware amplification, I don't want to (like many others) spend money on a simple little problem like this. Thanks!

  Read the article

• New Reference Configuration: Accelerate Deployment of Virtual Infrastructure

  - by monica.kumar

  Today, Oracle announced the availability of Oracle VM blade cluster reference configuration based on Sun servers, storage and Oracle VM software. Assembling and integrating software and hardware systems from different vendors can be a huge barrier to deploying virtualized infrastructures as it is often a complicated, time-consuming, risky and expensive process. Using this tested configuration can help reduce the time to configure and deploy a virtual infrastructure by up to 98% as compared to putting together multi-vendor configurations. Once ready, the infrastructure can be used to easily deploy enterprise applications in a matter of minutes to hours as opposed to days/weeks, by using Oracle VM Templates. Find out more: Press Release Business whitepaper Technical whitepaper

  Read the article

• Of these 3 methods for reading linked lists from shared memory, why is the 3rd fastest?

  - by Joseph Garvin

  I have a 'server' program that updates many linked lists in shared memory in response to external events. I want client programs to notice an update on any of the lists as quickly as possible (lowest latency). The server marks a linked list's node's state_ as FILLED once its data is filled in and its next pointer has been set to a valid location. Until then, its state_ is NOT_FILLED_YET. I am using memory barriers to make sure that clients don't see the state_ as FILLED before the data within is actually ready (and it seems to work, I never see corrupt data). Also, state_ is volatile to be sure the compiler doesn't lift the client's checking of it out of loops. Keeping the server code exactly the same, I've come up with 3 different methods for the client to scan the linked lists for changes. The question is: Why is the 3rd method fastest? Method 1: Round robin over all the linked lists (called 'channels') continuously, looking to see if any nodes have changed to 'FILLED': void method_one() { std::vector<Data*> channel_cursors; for(ChannelList::iterator i = channel_list.begin(); i != channel_list.end(); ++i) { Data* current_item = static_cast<Data*>(i->get(segment)->tail_.get(segment)); channel_cursors.push_back(current_item); } while(true) { for(std::size_t i = 0; i < channel_list.size(); ++i) { Data* current_item = channel_cursors[i]; ACQUIRE_MEMORY_BARRIER; if(current_item->state_ == NOT_FILLED_YET) { continue; } log_latency(current_item->tv_sec_, current_item->tv_usec_); channel_cursors[i] = static_cast<Data*>(current_item->next_.get(segment)); } } } Method 1 gave very low latency when then number of channels was small. But when the number of channels grew (250K+) it became very slow because of looping over all the channels. So I tried... Method 2: Give each linked list an ID. Keep a separate 'update list' to the side. Every time one of the linked lists is updated, push its ID on to the update list. Now we just need to monitor the single update list, and check the IDs we get from it. void method_two() { std::vector<Data*> channel_cursors; for(ChannelList::iterator i = channel_list.begin(); i != channel_list.end(); ++i) { Data* current_item = static_cast<Data*>(i->get(segment)->tail_.get(segment)); channel_cursors.push_back(current_item); } UpdateID* update_cursor = static_cast<UpdateID*>(update_channel.tail_.get(segment)); while(true) { if(update_cursor->state_ == NOT_FILLED_YET) { continue; } ::uint32_t update_id = update_cursor->list_id_; Data* current_item = channel_cursors[update_id]; if(current_item->state_ == NOT_FILLED_YET) { std::cerr << "This should never print." << std::endl; // it doesn't continue; } log_latency(current_item->tv_sec_, current_item->tv_usec_); channel_cursors[update_id] = static_cast<Data*>(current_item->next_.get(segment)); update_cursor = static_cast<UpdateID*>(update_cursor->next_.get(segment)); } } Method 2 gave TERRIBLE latency. Whereas Method 1 might give under 10us latency, Method 2 would inexplicably often given 8ms latency! Using gettimeofday it appears that the change in update_cursor-state_ was very slow to propogate from the server's view to the client's (I'm on a multicore box, so I assume the delay is due to cache). So I tried a hybrid approach... Method 3: Keep the update list. But loop over all the channels continuously, and within each iteration check if the update list has updated. If it has, go with the number pushed onto it. If it hasn't, check the channel we've currently iterated to. void method_three() { std::vector<Data*> channel_cursors; for(ChannelList::iterator i = channel_list.begin(); i != channel_list.end(); ++i) { Data* current_item = static_cast<Data*>(i->get(segment)->tail_.get(segment)); channel_cursors.push_back(current_item); } UpdateID* update_cursor = static_cast<UpdateID*>(update_channel.tail_.get(segment)); while(true) { for(std::size_t i = 0; i < channel_list.size(); ++i) { std::size_t idx = i; ACQUIRE_MEMORY_BARRIER; if(update_cursor->state_ != NOT_FILLED_YET) { //std::cerr << "Found via update" << std::endl; i--; idx = update_cursor->list_id_; update_cursor = static_cast<UpdateID*>(update_cursor->next_.get(segment)); } Data* current_item = channel_cursors[idx]; ACQUIRE_MEMORY_BARRIER; if(current_item->state_ == NOT_FILLED_YET) { continue; } found_an_update = true; log_latency(current_item->tv_sec_, current_item->tv_usec_); channel_cursors[idx] = static_cast<Data*>(current_item->next_.get(segment)); } } } The latency of this method was as good as Method 1, but scaled to large numbers of channels. The problem is, I have no clue why. Just to throw a wrench in things: if I uncomment the 'found via update' part, it prints between EVERY LATENCY LOG MESSAGE. Which means things are only ever found on the update list! So I don't understand how this method can be faster than method 2. The full, compilable code (requires GCC and boost-1.41) that generates random strings as test data is at: http://pastebin.com/e3HuL0nr

  Read the article

• Using Openfire for distributed XMPP-based video-chat

  - by Yitzhak

  I have been tasked with setting up a distributed video-chat system built on XMPP. Currently my setup looks like this: Openfire (XMPP server) + JingleNodes plugin for video chat OpenLDAP (LDAP server) for storing user information and allowing directory queries Kerberos server for authentication and passwords In testing with one set of machines (i.e. only three), everything works as expected: I can log in to Openfire and it looks up the user information in the OpenLDAP database, which in turn authenticates my user with Kerberos. Now, I want to have several clusters, so that there is a cluster on each continent. A typical cluster will probably contain 2-5 servers. Users logging in will be directed to the closest cluster based on geographical location. Something that concerns me particularly is the dynamic maintenance of contact lists. If a user is using a machine in Asia, for example, how would contact lists be updated around the world to reflect the current server he is using? How would that work with LDAP? Specific questions: How do I direct users based on geographical location? What is the best architecture for a cluster? -- would all traffic need to come into a load-balancer on each one, for example? How do I manage the update of contact lists across all these servers? In general, how do I go about setting this up? What are the pitfalls in doing this? I am inexperienced in this area, so any advice and suggestions would be appreciated.

  Read the article

• What is the role of web hosting in SEO [closed]

  - by Vinay

  Possible Duplicate: Does changing web hosting server affects SEO page ranking? SEO Geolocation What are the best ways to increase a site's position in Google? How to find web hosting that meets my requirements? I have read somewhere that hosting providers do play a role in website SEO, As my website is hosted in yahoo small business, That has got analytics and some other tool they provide to check the keyword activity, I think these can be achieved with the google's analytics as well, Server performance and Uptime is one important factor. I have also got few doubts in my mind 1) Does shared hosting affect the SEO and what is the role of domain extension like .com, .in, .org ,etc. 2) Does server geolocation affects the SEO 3) Does server OS affects the SEO. Apart from the above, Is there any factors that affect the SEO One more last question that If hosting really matters lot, can you suggest me a web hosting service for a small business e commerce site for PHP

  Read the article

• Develop website locally and push updates on Remote Server using Git

  - by John

  Together with a friend we are looking to develop a website (using Symfony2). We are on a Shared Hosting with SSH access. Below is the environment we'd like to setup: * Use git as Version Control (we are new to Git) * Share the tasks and develop on our local machines * Push the updates onto the remote server Here's our initial thoughts on how to do it (assuming Git is already running both locally and remotely): * Install Symfony on the Remote Server (basic setup) * Get a clone (using Git) of the project locally * Develop project locally and push updates (using Git) on the remote server Does this approach make sense, if not, any recommendations? Thanks

  Read the article

< Previous Page | 33 34 35 36 37 38 39 40 41 42 43 44  | Next Page >