Search Results

Search found 22065 results on 883 pages for 'performance testing'.

Page 711/883 | < Previous Page | 707 708 709 710 711 712 713 714 715 716 717 718  | Next Page >

• GRE Tunnel over IPsec with Loopback

  - by Alek

  I'm having a really hard time trying to estabilish a VPN connection using a GRE over IPsec tunnel. The problem is that it involves some sort of "loopback" connection which I don't understand -- let alone be able to configure --, and the only help I could find is related to configuring Cisco routers. My network is composed of a router and a single host running Debian Linux. My task is to create a GRE tunnel over an IPsec infrastructure, which is particularly intended to route multicast traffic between my network, which I am allowed to configure, and a remote network, for which I only bear a form containing some setup information (IP addresses and phase information for IPsec). For now it suffices to estabilish a communication between this single host and the remote network, but in the future it will be desirable for the traffic to be routed to other machines on my network. As I said this GRE tunnel involves a "loopback" connection which I have no idea of how to configure. From my previous understanding, a loopback connection is simply a local pseudo-device used mostly for testing purposes, but in this context it might be something more specific that I do not have the knowledge of. I have managed to properly estabilish the IPsec communication using racoon and ipsec-tools, and I believe I'm familiar with the creation of tunnels and addition of addresses to interfaces using ip, so the focus is on the GRE step. The worst part is that the remote peers do not respond to ping requests and the debugging of the general setup is very difficult due to the encrypted nature of the traffic. There are two pairs of IP addresses involved: one pair for the GRE tunnel peer-to-peer connection and one pair for the "loopback" part. There is also an IP range involved, which is supposed to be the final IP addresses for the hosts inside the VPN. My question is: how (or if) can this setup be done? Do I need some special software or another daemon, or does the Linux kernel handle every aspect of the GRE/IPsec tunneling? Please inform me if any extra information could be useful. Any help is greatly appreciated.

  Read the article

• Ubuntu 10.04 Server on Hyper-V Server R2 has sluggish install and command line

  - by Paul Hobart

  I've installed Ubuntu Server 10.04 (64 bit) on a Hyper-V Server R2. I've encountered two issues that I think are related: Very slow install Very slow command prompt The text-mode installer goes through a series of text-based prompt windows. It takes 7-10 seconds for each of these windows to draw on the screen. The end result is that every time I answer a prompt and hit enter I wait for 15 seconds while the screen redraws line by line. I can literally see each line of text being drawn (like the old 300 baud modems days). Once done installing, scrolling on the command line is super slow. For instance, if a simple command, like "ls", causes the screen to scroll, it will scroll very slowly. This happens on a fresh install. The server functions as a LAMP server and an OpenSSH server, but that's it (I don't even have any Virtual Hosts set up yet). AND this only happens on the Virtual Machine console. I access the console through Hyper-V Manager and don't have this problem on any of my other Virtual Machines. Also, this problem does NOT happen when accessing a shell through OpenSSH. How can I improve this performance issue?

  Read the article

• How can I bridge a VM to a remote network?

  - by asciiphil

  I have a system running QEMU/KVM (via libvirt). One of its VMs needs to have a presence on a subnet that is not local to the VM host. I have a Linux system on the remote subnet. Is there a way to set up some sort of tunneled bridge to cause the VM to appear present on the remote system? This will be a temporary situation (hopefully just until the VM owner can configure their system) and network performance and long-term maintainability aren't really issues. To give some more concrete information: My VM host has IP address 192.168.54.155/24. The VM has IP address 192.168.65.71/24. I have a remote system at 192.168.65.254/24. Both the VM host and remote system are running Scientific Linux 6.5. I do not control the network or routing in between the VM host and remote system. I do not have access to the guest OS on the VM. I would like traffic to the VM's IP address to end up at the VM even though its host isn't directly connected to the appropriate network. I've tried using iproute2's tunnelling, but Linux won't let me add a tunnel to a bridge. I've considered using some sort of iptables mangling to route traffic over the tunnel and make the VM think it's on the right network, but I'm not sure whether there are better approaches. What's the best way to accomplish this hack?

  Read the article

• How can I "share" a network share over the internet to multiple operating systems?

  - by Minsc

  Hello all, We have a network share accessible through our intranet that is widely used. This share has it's own set of fine tuned permissions. I have been tasked with allowing A.D. authenticated access to this share over the internet without the use of VPN. The internet access has to mimic the NTSF permissions in place on the share. Another piece of the puzzle is that the access over the internet has to allow perusal of the share from Windows and Mac OS systems. I had envisioned a web front end that would facilitate downloading to and uploading from the share via a web browser. I'm trying to ask for some suggestions about what type of setup is necessary to achieve this. I've done loads of testing and searching for solutions but I can't seem to get anything to work as I hope. The web server that will be handing all of this is a Windows 2K8 box with IIS 7. How can I allow the users to authenticate against Active Directory when coming from the internet even when coming from a Mac system? I hope my question is not too broad, I'm sorry if I should have broken it up into multiple questions. It all is just tied together in my head. Thank you all for your time and aid.

  Read the article

• Openfire on EC2 with Jingle

  - by Bjorn Roche

  I would like to run Openfire (or another XMPP server) on EC2. At the moment this is just for testing, so easy setup and configuration are important, as is low cost. At some point, however, if things go well, it will be important to scale this. Ideally, it would be nice to not have to switch software when the scaling happens, but if a switch needs to happen later it certainly can. My requirements are: basic XMPP services, including muc and pubsub. Logins controlled from an external API. Preferably, when a user attempts to connect, the XMPP server checks with the api to see if their username and password are correct, but I can also have the API keep the XMPP server up to date on new users, deleted users, pasword changes and so on. I see Openfire has a "user service" API. Not ideal, but it looks workable. Jingle, including relay and STUN. It's not at all clear to me if the Jingle Nodes plugin takes care of this. I'm a bit confused about what's required to set this up, and I'd rather know in advance than be confused along the way :). eg It seems like STUN servers require more than one IP address. Can Openfire do all this for me, including stun and media relay on a single machine? Is this hard to configure on EC2 with Openfire? What are the basic steps? Would this be easier with something else like, say Tigase? What about database? Should I use amazon's database service, or run a db on the same machine? Would the server be compatible with a service like http://www.siteuptime.com/ Thanks!

  Read the article

• Using bind (named) as a public proxy server

  - by TrentDavis

  We have a Python DNS server that does a bunch of stuff to figure out values it should return for various DNS records. This works nicely, however as it is Python, the performance under high load won't be great. What I would like to do is have a "proxy" bind server sit in front of it to return results to the public internet. This will cache the results (typically 15 minutes, some records are a few seconds), so the load on the Python server will be greatly reduced as it will only see one look up per domain (only about 100 domains) every 15 minutes. The data in these domains changes a lot, so using a master won't work as it will constantly be changing. I have something setup that looked like it would work great (using a forwarder for the zone), and tested it with dig etc, all going great. However when we went to go live with it, things weren't working, and we figured out that named is not setting the "Authoritative" bit (fair enough, it is a forwarder). So my question is, can we tell bind to set the Authoritative bit for forwarded domains? I have looked at all the doco I can find, and can't find anything about doing things this way. Most of the doco about using it as a proxy if for a LAN to the internet. Ideally I would like to use bind as it is there and installed (CentOS 5 servers). But at a pinch we could look at a different name server to do the work if it just can't be done with bind. Thanks.

  Read the article

• SharePoint Backup/Restore without stsadm

  - by Kevin

  Due to problems we found with the restore of sites/site collections using stsadm (our tasks generated from workflows were not restored), we've taken a different route for backup/restore. We plan a major customization to our SP site and want to take a backup so we can rollback in case the install fails. In our System Testing (not production) environment, we've backed up the 12 hive, the virtual dir's that the IIS points to SharePoint, and the SharePoint databases in SQL (using SQL server to do the db backups). We have custom event handlers and workflows built with Visual Studio, and deploy the dlls to the GAC as version 2 (signed and versioned in Visual Studio). So when we deploy, the GAC will contain 2 versions of the workflows - version 1 and version 2. During the deploy we use SP stsadm features to install/activate the WF's. We also go to each library and add the new, version 2 WFs. This automatically sets the version 1 WF's to "Not Allow" new instances (which is what we want) and the version 2 as active - perfect so far. When we've completed the install, we then assume a failure and attempt to restore to the same machines (SharePoint on one server, SQL on another). We start by uninstalling the version 2 WF's from the GAC, reset IIS (to clear cache of these ver. 2 WF dlls'), restore the 12-hive and virtual directory folders, then restore the SQL dbs. This is all just as manual as you read it - no stsadm here. All seems to work after our restore, it appears the restore was successful - the mods we made to column names, data changes, etc during the install are all reverted back to the original pre-install state. With one exception. When we run a workflow, it always fails and the Logs in the 12-hive indicates the WF is still trying to use the version 2 of the dll (System.IO file not found error) We think we've backed up and restored all the moving pieces of Sharepoint but we're missing something here, does anybody have any ideas why the version 2 WF dlls are still being referenced eventhough we restored all the folders and db's of SharePoint? Thanks, Kevin

  Read the article

• SQL Server on Linux

  - by TimothyAWiseman

  For a particular project that is coming up, I am trying to expand my knowledge of Linux, so I am going to set up a Linux system at home. Rather than dual booting, I am thinking about putting SQL Server on a Windows Virtual Machine with Linux as the host at least until this project is over when I will probably switch back to Linux. So, I have a couple of different, but interrelated questions: How well does this work? This is only a test machine at home, so I can easily accept a fair bit of degradation, but if it is going to be a horrible reduction in performance I will dual boot instead. Is there a particular virtual machine manager I should look at to go this route? Since this is my personal machine, price is an issue but I am quite happy to pay a reasonable amount. And finally, given the choice of VMM, is there a particular Linux Distro I should be looking at? [This has been cross posted at Ask.SqlServerCentral.com . I think it may be appropriate at both sites. ]

  Read the article

• Mod_Perl configuration for multiple domains

  - by daliaessam

  Reading the Mod_Perl module documentation, can we configure it on per domain basis, what I mean can we configure it to run on every domain or specific domain only. What I see in the docs is: Registry Scripts To enable registry scripts add to httpd.conf: Alias /perl/ /home/httpd/2.0/perl/ <Location /perl/> SetHandler perl-script PerlResponseHandler ModPerl::Registry PerlOptions +ParseHeaders Options +ExecCGI </Location> and now assuming that we have the following script: #!/usr/bin/perl print "Content-type: text/plain\n\n"; print "mod_perl 2.0 rocks!\n"; saved in /home/httpd/httpd-2.0/perl/rock.pl. Make the script executable and readable by everybody: % chmod a+rx /home/httpd/httpd-2.0/perl/rock.pl Of course the path to the script should be readable by the server too. In the real world you probably want to have a tighter permissions, but for the purpose of testing, that things are working, this is just fine. From what I understand above, we can run Perl scripts only from one specific folder that we put the directive above. So the question again, can we make this directive per domain for all domains or for specific number of domains?

  Read the article

• how to cause linux system datetime to run faster than real world datetime?

  - by JamesThomasMoon1979

  Background I want to monitor a running linux system over several days. It's a custom gentoo build and with much custom software on board. This software has ongoing maintenance timers and cron scripts and other clock driven events. I need to verify these scheduled events are working. Problem Waiting for the system to step through daily and weekly activity is a long wait time. And modifying all clock-based timers on the system would be time consuming. Yet, I often want to test a system's end-to-end scheduled activities without waiting a week. Potential Solution Have the linux system under test appear to run through it's daily cycle of activity within just a few hours. My Question for Serverfault Is there a way to cause the system's time to run faster than real world time? My first thought is manipulating the ntp daemon to repeatedly and smoothly increment the clock . Any other ideas? And yes, I know this may have strange side affects. However, the system has no important or time critical interactions with systems outside of itself. And this may be a valuable testing technique.

  Read the article

• mysql mass insert data

  - by user12145

  Edit: I realized that if I construct a large query in memory, the speed has increased almost 10 times of magnitude "insert ignore into xxx(col1, col2) values('a',1), values('b',1), values('c',1)..." Edit: since I have an index on the first column, the insert time creeps up as I insert more. Can I delay the index until the end? Original: I'm using the following to batch insert 10 million rows into mysql db(not all at once, since they don't all fit into memory), it's too slow(taking many hours). should I use load file to improve performance? I would have to create a second file to store all the 10 million rows, then load that into db. are there better ways? PreparedStatement st=con.prepareStatement("insert ignore into xxx (col1, col2) "+ " values (?, 1)"); Iterator d=data.iterator(); while(d.hasNext()){ st.clearParameters(); st.setString(1, (d.next()).toLowerCase()); st.addBatch(); } int[]updateCounts=st.executeBatch();

  Read the article

• OS X mouse pointer speed varies with different mouse

  - by Stan

  OS X Snow Leopard It seems that when using different mice on OS X may have different pointer speed and scrolling speed. For example, when using my Logitech basic laser mouse, the pointer speed is like normal. But when using MX Performance or Anywhere, it's very slow, I will have to adjust the pointer speed in mouse configuration to max. Even with max, it's still a bit slow. Basically, just feel the plug and play on OS X is terrible. I need re-adapt to it every single time. This is not the case on Windows OS. Also, the mouse scrolling speed varies with different mouse too. But usually they are all very slow, usually scroll 1 line at a time. If I adjust it in mouse configuration, it turns to scroll too much lines. I have Logitech official mouse driver (LCC) installed. But either tuning in LCC or mouse configuration doesn't make things better. Has anyone have similar issue? How to resolve it? Please advise, thanks.

  Read the article

• What benchmark tool to use to benchmark hardware for VM server?

  - by Mark0978

  We are setting up a new piece of hardware to virtualize several of our servers on. Choices are RAID 5, RAID 6, and RAID 0+1. We are wanting to benchmark all three before we go live with the machine, but I'm not sure how to test the speed. Since we will be using it to host VMs, what will the actual disk traffic look like? What can I use to see if RAID 6 is too slow? Short of setting up the system with all the VM's on it and running that way, then redoing on all the work, I'm not sure how to test it. It them becomes more of a subjective test than an objective one. I'm worried that RAID6 will have too much overhead, that RAID5 will be to fragile with 3TB drives and I've never worked with 0+1 at all. So in short I'd like to setup the base machine (which will be running Linux) and then test the underlying SW RAID for speed. What kind of tool exists to simulate this kind of load? Barring the lack of a specific tool, how about a generic FS testing tool that will simulate different loads?

  Read the article

• How much network latency is "typical" for east - west coast USA?

  - by Jeff Atwood

  At the moment we're trying to decide whether to move our datacenter from the west coast to the east coast. However, I am seeing some disturbing latency numbers from my west coast location to the east coast. Here's a sample result, retrieving a small .png logo file in Google Chrome and using the dev tools to see how long the request takes: West coast to east coast: 215 ms latency, 46 ms transfer time, 261 ms total West coast to west coast: 114 ms latency, 41 ms transfer time, 155 ms total It makes sense that Corvallis, OR is geographically closer to my location in Berkeley, CA so I expect the connection to be a bit faster.. but I'm seeing an increase in latency of +100ms when I perform the same test to the NYC server. That seems .. excessive to me. Particularly since the time spent transferring the actual data only increased 10%, yet the latency increased 100%! That feels... wrong... to me. I found a few links here that were helpful (through Google no less!) ... Does routing distance affect performance significantly? How does geography affect network latency? Latency in Internet connections from Europe to USA ... but nothing authoritative. So, is this normal? It doesn't feel normal. What is the "typical" latency I should expect when moving network packets from the east coast <--> west coast of the USA?

  Read the article

• Restoring snapshot for Microsoft Exchange server

  - by Mugen

  Hi, The background: I need to do some testing with Microsoft exchange server. Specifically, I'll be installing some software on the Microsoft exchange server machine and uninstalling that same software again. The problem I face: While I repeatedly do this with different versions of my software there is a chance that sometime later the Exchange server installation might get corrupted. When that happens I would need to reinstall Exchange server which I feel is somewhat of a chore. So what I am planning to do is to install the Ms Exchange server on a virtual machine in VMware ESX server and take a snapshot so that during my work whenever the installation is corrupted I can restore the snapshot. So here's my question: Would restoring the snapshot for the Microsoft Exchange server virtual machine work correctly? I'm not familiar with the intricacies of exchange server and any changes (if any) that happen with the Domain controller when we use or install an exchange server (Personally I don't think that should happen but just making sure). I have a shortage of time and hence decided to post this question here. Could someone please tell me whether restoring a snapshot for Exchange server would work fine? Thanks a load, Mugen

  Read the article

• hdparm - how to secure erase SATA SSD over USB

  - by cc0

  I have been following this guide on how to secure erase an SSD (trying to improve the performance of mine, which currently only writes at about 30mb/s seq). However, I'm using an USB--Sata docking device to avoid having the harddrive frozen. Apparently using this solution the SATA device is recognized as a SCSI drive, which is giving me trouble. I use the "hdparm -I /dev/sda" command with those parameters, and I get the error; HDIO_DRIVE_CMD (identify) failed: Invalid Exchange After a lot of googling on the issue I can't seem to find anyone who has actually solved this problem. However, I have not tried to just go ahead and use the secure erase. So I'm not sure if this would actually still work. I would love any and all input I can get on this, especially on whether it will still work to do a secure erase with the drive being recognized as a SCSI drive. The drive itself is a Samsung 256gb SSD (pm800), I'm sure you can understand my reluctance to go through this procedure without feeling reasonably safe that I won't mess it up beyond repair.

  Read the article

• Windows Server 2008 32 bit & windows 7 professional SP1

  - by Harry

  I'm testing my new Windows Server 2008 32 bit edition (2 servers) as a server and Windows 7 professional 32 bit as a client. Let say one is a primary domain controller (PDC) and the other is a backup domain controller (BDC) like the old time to ease. Every setup were done in the PDC and just replicate to BDC. Didn't setup anything, just install the server with AD, DNS, DHCP, that's all. Then I use my windows 7 pro 32 bit to join the domain. It worked. After that I tried to change the password of a the user (not administrator) but it always failed said it didn't meet the password complexity setup while in fact there's no setup at all either in account policy, default domain policy or even local policy. Tried to disable the password complexity in the default domain policy instead of didn't set all then test again but still failed. Browse and found suggestion to setup the minimum and maximum password age to 0 but it also failed. Tried to restart the server and the client then change password, still failed with the same error, didn't meet password complexity setup. Tried to see in the rsop.msc but didn't found anything. In fact, if I see the setup in another system with windows server 2003 and windows xp, using rsop.msc I can see there's setup for computer configuration windows settings security settings account policies password policy. I also have a windows 7 pro 32 bit in a windows server 2003 32 bit environment but unable to find the same setting using rsop but this windows 7 works fine. anyone can give suggestion what's the problem and what to do so I can change my windows 7 pro laptop password in a windows server 2008 environment? another thing, is it the right assumption that we can see all the policies setting in windows 7 whether it's in a windows server 2003 or 2008 environment? thanks.

  Read the article

• SQL Server Windows-only Authentication Strategy problem

  - by Mike Thien

  I would like to use Windows-only Authentication in SQL Server for our web applications. In the past we've always created the all powerful 1 SQL Login for the web application. After doing some initial testing we've decided to create Windows Active Directory groups that mimic the security roles of the application (i.e. Administrators, Managers, Users/Operators, etc...) We've created mapped logins in SQL Server to these groups and given them access to the database for the application. In addition, we've created SQL Server database roles and assigned each group the appropriate role. This is working great. My issue revolves around that for most of the applications, everyone in the company should have read access to the reports (and hence the data). As far as I can tell, I have 2 options: 1) Create a read-only/viewer AD group and put everyone in it. 2) Use the "domain\domain users" group(s) and assign them the correct roles in SQL. What is the best and/or easiest way to allow everyone read access to specific database objects using a Windows-only Authentication method?

  Read the article

• Domain authentication over OPEN wireless pre-logon (Windows 7 Pro) - No logon servers avail

  - by Shadow00Caster

  I have a plethora of laptops that are joined to an AD domain. I have an enterprise wireless system setup, the users of these laptops will be using an OPEN unsecured SSID which will ultimately have a captive portal that uses Radius-AD auth and firewall rules to allow access pre-captive portal auth to the proper ip's/ports of DC's etc for auth etc. I already have other laptops/users connecting to another SSID with 802.11x and SSO, all works perfectly pre-logon etc. My problem is with this open network, for some reason I cannot get the machines to auth to AD. The laptops connect to the wireless network, I confirm this on the controller and can ping the laptop at startup. I sharked the wires on the 2 DC's that these machines auth to, I can see a DNS SOA update from a laptop im testing with and can ping that test laptop from both DC's. When I try to logon, "There are currently no logon servers available to service the logon request." The shark shows no incoming connections to either DC even though the laptop is connected and pingable. Any help is greatly appreciated.

  Read the article

• Issues with sustained traffic with PFSense

  - by Farseeker

  Last week we had to replace our PFSense firewall because it had a catastrophic hardware failure. All but one of the NICs were taken out of the old server and put into the new one. The one NIC that was not moved was the LAN NIC as this is on-board. The other NICs are all WAN connections and the must all be present (i.e. I can't disable one just for the sake of testing) After re-installing PFSense and restoring our backup of the configuration, everything came back online just fine, however on the new hardware any download that takes longer than about 10 seconds just times out in the middle. Example 1: Downloading from Microsoft.com goes at about 900k/sec and times out after about 10 seconds (thus, just under 10Mb of content) Example 2: Downloading from cnet.com goes at about 300k/sec and times out after about 10 seconds (thus, about 3Mb of content). By times out, I mean that the download just stops, and you have to pause/resume to get the next part done, repeat and rinse until the download is complete. However it's not consistant, sometimes it's 10 seconds, sometimes it's 4 seconds, and it sometimes you can't even load a heavy HTML page because the page never finishes. I assume this is most likely because PFSense does not like the onboard NIC, as this is the primary difference between the two servers. It's recognised as NFE0, and there's no room in the server for any more NICs and I don't have any dual-port NICs handy to experiment with a different LAN connection. I've never had to troubleshoot this sort of issue before. Can anyone give me some pointers about where to start? Linux is not my forte so please be kind!

  Read the article

• Why is squid breaking kerberos/NTLM auth?

  - by DonEstefan

  I'm using squid 2.6.22 (Centos 5 Default) as a proxy. Squid seems to break the authentication process for web pages when they require NTLM or Kerberos Auth. I tested with sharepoint 2007 and tried all 3 authentication methods (NTLM, Kerberos, Basic). Accessing the site without squid works in all cases. When I access the same page with squid, then only basic-auth works. Using IE or Firefox desn't make any difference. Squid itself can be used by anybody (no auth_param configured). Its a bit tricky to find solutions online, since most of the topics whirl around auth_param for authenticating users to squid rather than authenticating users to a webpage behind squid. Could anyone help? Edit: Sorry, but my first test was totally screwed up. I tested against the wrong webservers (Memo to myself: always check assumptions before testing). Now I realized that the problem scenario is completely different. Kerberos work for IE Kerberos works for Firefox (after changing "network.negotiate-auth.trusted-uris" in about:config) NTLM works for IE NTLM does NOT work in Firefox (even after changing "network.automatic-ntlm-auth.trusted-uris" in about:config) By the way: The feature that provides NTLM-passthrough in squid is called "connection pinning" and the HTTP header "Proxy-support: Session-based-authentication""

  Read the article

• SharePoint Records Center Submitted E-mail Records not picked up

  - by Kenneth Verburg

  We have set up a new SharePoint 2007 site with a Records Repository. We're using Exchange 2007 Managed Folders to route e-mails to this repository based on the 'label' attached to the e-mail as set in the Exchange 2007 journaling options. E-mails added to a Managed Folder get sent to SharePoint, they end up in the "Submitted E-mail Records" list of the Records Repository. That's according to plan, but the e-mails are not routed to the respective document library as defined by the label. Instead an error appears in the event viewer for every e-mail listed in the Submitted E-mail Records list, on every interval of the records repository schedule (set to every two minutes for testing purposes): Value cannot be null, parameter name: g. Sending a document from the SharePoint site iself to the Records Repository via the Send To... link works fine, but e-mails get stuck in the list... We have set Document Libraries in the Respository with and without content types (with matching names with the Label and the Record Routing rule set). Any ideas what could be wrong? This is in the event log: Every two minutes the following error appears in the Application Log: Source: Office SharePoint Server Category: Records Center Type: Error Event ID: 4975 User: N/A Computer: SPS2007 Description: Value cannot be null. Parameter name: g For more information, see Help and Support Center at http://go.microsoft.com/fwlink/events.asp.

  Read the article

• btrfs: can i create a btrfs file system with data as jbod and metadata mirrored

  - by Yogi

  I am trying to build a home server that will be my NAS/Media server as well a the XBMC front end. I am planning on using Ubuntu with btrfs for the NAS part of it. The current setup consists of 1TB hdd for the OS etc and two 2TB hdd's for data. I plan to have the 2TB hdd's used as JBOD btrfs system in which i can add hdd's as needed later, basically growing the filesystem online. They way I had setup the file system for testing was while installing the OS just have one of the HDD's connected and have btrfs on it mounted as /data. Later on add another hdd to this file system. When the second disk was added btrfs made as RAID 0, with metadata being RAID 1. However, this presents a problem: even if one of the disk fails I loose all my data (mostly media). Also most of the time the server will be running without doing any disk access, i.e. the HDD's can be spun down, when a access request comes in this with the current RAID 0 setup both disks will spin up. in case I manage a JBOD only the disk that has the file needs to be spun up. This should hopefully reduce the MTBF for each disk. So, is there a way in which I can have btrfs setup such that metadata is mirrored but data stays in a JBOD formation? Another question I have is this, I understand that a full drive failure in JBOD will lose data on the drive, but having metadeta mirrored across all drives, will this help the filesytem correct errors that migh creep in (ex bit rot?) and is btrfs capable of doing this.

  Read the article

• Solaris Fibre Channel target - Configure QLogic QLA2340

  - by growse

  I'm currently trying to set up a small storage system as a fibre channel target. This is for testing, so I'm currently using Solaris (Nexenta) and a QLogic QLA2340 HBA. For some reason, the qlc and qlt drivers don't support the QLA2340, so I'm using the qla2300 driver from QLogic's website. I've also got the scli utility installed for configuration. The HBA is detected by the system. That said, it's not clear how I get from this point to a point where I have a ZFS volume being exposed as an FC target. I was originally following this guide (http://www.youtube.com/watch?v=yzEBd3l7Qn4) but it seems that without the qlc/qlt drivers, Sun's configuration tools won't work. Does that also imply that COMSTAR also won't work? What's the best way to expose an FC target with this setup? Most of the options I'm seeing in scli complain that the port state is LinkDown (it is, I've not plugged anything in yet). Do I have to have my FC client plugged up and working before I can configure the target? Apologies for the slight vagueness of the question, but I'm not overly familiar with the terminology.

  Read the article

• Intel Rapid Storage Technology (pre-OS) driver installation

  - by Nero theZero

  My desktop machine is built on Gigabyte GA-Z87-UD3H and Gigabyte provides the latest driver for Intel Rapid Storage Technology (IRST), which I installed after installing the OS. Same goes for my Lenovo Thinkpad-T420. And for both machine, checking the controller device under the IDE ATA/ATAPI Controllers section in Device Manager I see the driver has been updated to the latest version. I set the SATA controller to AHCI from BIOS On the desktop machine I have one WD 2TB BLACK & one WD 3TB Green I don’t use RAID, & no chance of using in near future, but according to Intel IRST improves performance in single disk scenario too. Now I have the following questions – What is the actual purpose of IRST (pre-OS install) driver that doesn’t get served with a post-OS driver that I installed? There must be some difference, otherwise there wouldn’t be a pre-OS version of the driver. Right? In the pre-OS procedure (loading the drivers at OS-installation time) after successfully completing the OS installation, do I need that post-OS driver? Because after installing from that one I got a quick launch icon that runs the IRST configuration application. Where do get that after installing the pre-OS driver? As it is “pre-OS”, when I load it at OS-installation time, does it updates anything at BIOS level or anywhere other than HDD? That’s because I’m going to dual boot Windows 7 with Windows 8.1, and after installing Windows 7 when I install Windows 8.1 & load the IRST driver for that, is there any chance of any “overwriting” or OS-incompatibility? In short, is there anything specific to follow while installing the second OS?

  Read the article

< Previous Page | 707 708 709 710 711 712 713 714 715 716 717 718  | Next Page >