Search Results

Search found 17314 results on 693 pages for 'vpn setup'.

Page 567/693 | < Previous Page | 563 564 565 566 567 568 569 570 571 572 573 574  | Next Page >

  • Find routers IP address on the other side

    - by corsiKa
    Here's the basic setup of my network In this diagram: 1: The internet c: cable 2: Wireless router w: wireless connection 3: A win7 box with internet connection sharing enabled 4: A wireless router, but I'm only using its LAN capabilities to connect box 5 to the internet. 5: A win7 box, the computer I'm using to make this post. So its internet works just fine. Now if I'm on box 5, and I ping 192.168.1.1, I hit 4. If I'm on box 3 and I ping 192.168.1.1, I hit 2. Now obviously box 3 does not think 4's IP address is 192.168.1.1, or I wouldn't be able to connect to the internet. Okay, now that you know as much as I do about my network, here's my question: If I was on box 3, how would I determine the IP address of 4? Basically I'm running a webserver on box 5 and want to access this webserver on box 3 and other boxes. So that's the end goal. If there's other information there that can help, I'd appreciate it. Thanks!

    Read the article

  • Hosting multiple websites from home

    - by dean nolan
    I have just been accepted for Microsofts Wevsite Spark program which I mainly got for the tools, Visual Studio, Blend. I also have a few of my own websites, personal and a couple of business ones. I also work freelance and sometimes I would like a place to just put a demo up of a clients project. The websites I currently have are all on differnet hosting provders and domain registrars. The WebsiteSpark comes with Windows Server 2008 and SQL Server 2008. It would be really advantagous of me to have all these in one place but also so I have complete control over the database and the environment. So I am thinking over the next 4-6 months of migrating all this to my own server that I will host from home, or maybe even setup at home and then store in a proper datacentre. I was wondering what steps I should take and what to be aware of, specifically: 1) having all these different websites on one computer and having the url got to the proper place. 2) Cost effectiveness? Having the server in home as apposed to datacentre. Most solutions I see charge over £1000 a month to have a machine in datacentre. This is mostly for my own ease of management and shared hosting which I currently have is very limited configuration wise. Would getting a server in house be beneficial for then upgrading to the cloud? What measures should I take with my ISP? I know this is a lot I've asked but just even links to good articles would be good. Thanks

    Read the article

  • IBM BladeCenter S: Disk Configuration

    - by gravyface
    Have just the one storage bay right now (SAS 15K 600GB x 6) and have configured one storage pool in RAID 10 with 4 disks (and two global spares). For each blade, I've created a volume and mapped accordingly: Blade #1 400 GB Blade #2 200 GB Blade #3 100 GB Blade #4 100 GB When I boot up Blade 1 and enter into the UEFI Setup (F1) followed by the Adapters and UEFI Drivers LSI Logic Fusion MPT SAS Driver Utility, I see 4 disks: two are the on-board 73GB drives, the other two are 200GB each and assume I'm being presented with two logical disks from the volume I created and mapped to this blade. I was a bit surprised by this: I figured I would've been presented with one logical drive per volume, not two. I'm assuming I can just configure whatever RAID level I wish that supports two disks, but not really sure what the benefits/trade-offs here. Should I go with RAID 10 on top of RAID 10? RAID 0? Software RAID 0/1/10? Does it even matter? If this is "normal" to see two disks, then I'm going to likely just do some benchmarking and see if it makes a difference changing the RAID levels (my guess is no); if this is not normal, well, please let me know. :)

    Read the article

  • System Center 2012 VMM UI is very slow

    - by Grant
    I've recently setup system center 2012 a new server 2008 r2 server which I'm using for virtual machines. Everything seems to be working fine, and the virtual machines are nice and fast. But the Virtual Machine Manager interface is always excruciatingly slow. Sometimes taking up to 15 seconds moving between screens. It's very frustrating trying to use it when a task that just involves a couple clicks ends up taking several minutes. Pages that have a lot of form fields seem to take the longest to load - such as the page to change hardware settings of a virtual machine. Is this just normal performance for VMM? If not, where can I look to find what is slowing it down. Nothing else on the system seems to suffer. I can load and use Hyper-V manager with no noticable slowness. Even programs like event viewer that are usually rather slow seem to load fairly fast. Only the system center programs seem slow. Server is a Dell R710, 2x16 core opteron 6274 processors, 96GB RAM. OS drive is 2x500GB 7.2k RPM SAS drives in RAID1 (opted for the less expensive 7.2k drives since pretty much everything is stored on the SAN). Am I just being impatient? Does anyone else use VMM 2012 and find it slow?

    Read the article

  • OpenVPN - client-to-client traffic working in one direction but not the other

    - by Pawz
    I have the following VPN configuration: +------------+ +------------+ +------------+ | outpost |----------------| kino |----------------| guchuko | +------------+ +------------+ +------------+ OS: FreeBSD 6.2 OS: Gentoo 2.6.32 OS: Gentoo 2.6.33.3 Keyname: client3 Keyname: server Keyname: client1 eth0: 10.0.1.254 eth0: 203.x.x.x eth0: 192.168.0.6 tun0: 192.168.150.18 tun0: 192.168.150.1 tun0: 192.168.150.10 P-t-P: 192.166.150.17 P-t-P: 192.168.150.2 P-t-P: 192.168.150.9 Kino is the server and has client-to-client enabled. I am using "fragment 1400" and "mssfix" on all three machines. An mtu-test on both connections is successful. All three machines have ip forwarding enabled, by this on the gentoo boxes: net.ipv4.conf.all.forwarding = 1 And this on the FreeBSD box: net.inet.ip.forwarding: 1 In the server's "ccd" directory is the following files: client1: iroute 192.168.0.0 255.255.255.0 client3: iroute 10.0.1.0 255.255.255.0 The server config has these routes configured: push "route 192.168.0.0 255.255.255.0" push "route 10.0.1.0 255.255.255.0" route 192.168.0.0 255.255.255.0 route 10.0.1.0 255.255.255.0 Kino's routing table looks like this: 192.168.150.0 192.168.150.2 255.255.255.0 UG 0 0 0 tun0 10.0.1.0 192.168.150.2 255.255.255.0 UG 0 0 0 tun0 192.168.0.0 192.168.150.2 255.255.255.0 UG 0 0 0 tun0 192.168.150.2 0.0.0.0 255.255.255.255 UH 0 0 0 tun0 Outpost's like this: 192.168.150 192.168.150.17 UGS 0 17 tun0 192.168.0 192.168.150.17 UGS 0 2 tun0 192.168.150.17 192.168.150.18 UH 3 0 tun0 And Guchuko's like this: 192.168.150.0 192.168.150.9 255.255.255.0 UG 0 0 0 tun0 10.0.1.0 192.168.150.9 255.255.255.0 UG 0 0 0 tun0 192.168.150.9 0.0.0.0 255.255.255.255 UH 0 0 0 tun0 Now, the tests. Pings from Guchuko to Outpost's LAN IP work OK, as does the reverse - pings from Outpost to Guchuko's LAN IP. However... Pings from Outpost, to a machine on Guchuko's LAN work fine: .(( root@outpost )). (( 06:39 PM )) :: ~ :: # ping 192.168.0.3 PING 192.168.0.3 (192.168.0.3): 56 data bytes 64 bytes from 192.168.0.3: icmp_seq=0 ttl=63 time=462.641 ms 64 bytes from 192.168.0.3: icmp_seq=1 ttl=63 time=557.909 ms But a ping from Guchuko, to a machine on Outpost's LAN does not: .(( root@guchuko )). (( 06:43 PM )) :: ~ :: # ping 10.0.1.253 PING 10.0.1.253 (10.0.1.253) 56(84) bytes of data. --- 10.0.1.253 ping statistics --- 3 packets transmitted, 0 received, 100% packet loss, time 2000ms Guchuko's tcpdump of tun0 shows: 18:46:27.716931 IP 192.168.150.10 > 10.0.1.253: ICMP echo request, id 63009, seq 1, length 64 18:46:28.716715 IP 192.168.150.10 > 10.0.1.253: ICMP echo request, id 63009, seq 2, length 64 18:46:29.716714 IP 192.168.150.10 > 10.0.1.253: ICMP echo request, id 63009, seq 3, length 64 Outpost's tcpdump on tun0 shows: 18:44:00.333341 IP 192.168.150.10 > 10.0.1.253: ICMP echo request, id 63009, seq 3, length 64 18:44:01.334073 IP 192.168.150.10 > 10.0.1.253: ICMP echo request, id 63009, seq 4, length 64 18:44:02.331849 IP 192.168.150.10 > 10.0.1.253: ICMP echo request, id 63009, seq 5, length 64 So Outpost is receiving the ICMP request destined for the machine on it's subnet, but appears not be forwarding it. Outpost has gateway_enable="YES" in its rc.conf which correctly sets net.inet.ip.forwarding to 1 as mentioned earlier. As far as I know, that's all that's required to make a FreeBSD box forward packets between interfaces. Is there something else I could be forgetting ? FWIW, pinging 10.0.1.253 from Kino has the same result - the traffic does not get forwarded. UPDATE: I've found that I can only ping certain IP's on Guchuko's LAN from Outpost. From Outpost I can ping 192.168.0.3 and 192.168.0.2, but 192.168.99 and 192.168.0.4 are unreachable. The same tcpdump behavior can be seen. I think this means the problem can't be due to ipforwarding or routing, because Outpost can reach SOME hosts on Guchuko's LAN but not others and likewise, Guchuko can reach two hosts on Outpost's LAN, but not others. This baffles me.

    Read the article

  • Automated Syslog Error Solution Finder

    - by Dru
    Any automated syslog solution finding frameworks? I want my central syslog server to email a list of problems, their severity and suggested solutions. There have been several questions about centralising system logs and alternative log analysis systems, but I don't get the impression that any of them help with issue resolution. A little background: At work I am now literally doing the work of two people, and both jobs have expanded beyond their initial frameworks. It is not so bad as I have helpers, but they are little more than smart monkeys. While one of my predecessors [I have two, that is how I know I have the jobs of two people] set-up logwatch to email its results out, my monkeys don't have the skills necessary to identify unimportant data. This has caused all of them, and myself sadly, to setup email filters and ignore the whole thing until something goes "bang". It would be handy to have someone else tell them what is important, what is connected, and to suggest a few ways to resolve the issue (I could train then to research the solution first, ha!). My reading of the Splunk and Octopussy sites indicates that I still need to bring my own highly trained monkey to the party. Which I am several years from having.

    Read the article

  • Remote access to phpmyadmin from computer belongs to same LAN

    - by Charles
    OK... I solved it. It is because I have not configured the httpd.conf to allow the centos listen port 80 and 8080. Listen 80 Listen 8080 I have setup the myphpadmin on my CentOS 6.4 recently. I can access and login to the myphpadmin on my localhost. However, when I type http://[hostipaddr]/phpmyadmin on my other computer in the same LAN with the CentOS, the browser simply cannot access the page. Below are some of the current configuration. Anyone can help please......? config.inc.php $i++; /* Authentication type */ $cfg['Servers'][$i]['auth_type'] = 'http'; /* Server parameters */ $cfg['Servers'][$i]['host'] = 'localhost'; $cfg['Servers'][$i]['connect_type'] = 'tcp'; $cfg['Servers'][$i]['compress'] = false; /* Select mysql if your server does not have mysqli */ $cfg['Servers'][$i]['extension'] = 'mysql'; $cfg['Servers'][$i]['AllowNoPassword'] = false; phpmyadmin.conf <Directory /var/www/html/phpmyadmin/> order allow,deny allow from all </Directory> Furthermore, I can access the webpage that stored in the CentOS from my other computer without problems. After using wireshark and tcpdump, I found that the server (the Cent OS) keep resetting the connection. (192.168.1.106 is my other computer, 192.168.1.101 is my CentOS) 23:29:42.281473 IP 192.168.1.106.55999 > 192.168.1.101.webcache: Flags [S], seq 2559409090, win 65535, options [mss 1460,nop,wscale 8,nop,nop,sackOK], length 0 23:29:42.281504 IP 192.168.1.101.webcache > 192.168.1.106.55999: Flags [R.], seq 0, ack 2559409091, win 0, length 0 I have disabled the iptables service on the CentOS already.

    Read the article

  • Very Slow DSL (ethernet) speed [New Interesting Update]

    - by Abhijit
    Very IMPORTANT and INTERESTING UPDATE: Due to some reason I just thought to do a complete new setup and this time I decided to again have openSUSE plus ubuntu. So I first reinstall lubuntu and then I installed OpenSUSE 12.2 (64 bit). Now, my DSL speed is working very normal and fine on opensuse. So this is very scary. Is it possible for any operating system to manipulate my NIC so that it will work fine only on that operating system and not on another os? Regarding positive thinking and not being paranoid, what is it that makes ONLY suse to get my NIC to work at normal speed but ubuntu can not do it? Not even fedora? Not even linux mint? What all these OS are lacking that enables suse to work great? == ORIGINAL QUESTION == I 'was' on opensuse 12.2 when my dsl speed was normal. Yesterday I switched from opensuse to ubuntu 12.04 and speed decreased. It came to range of 7-10-13-20-25-kbps. Then I switch to linux mint, and then to fedora. Still slow speed. When I was in ubuntu I disabled ipv6 but still no luck. Now I am in fedora but this time with DIFFERENT ISP. And still I am getting very slow sped. So my guess is this is nothing to do with os. What can be wrong? Is this problem of NIC? Does NIC speed decreases over time? Does NIC life ends over time as with keyboard or mouse? Help please All the os I used are 64 bit and my laptop is Compaq Presario A965Tu Intel Centrino DUal Core. Interesting thing to notice is I get normal speed while downloading torrent inside torrent client softwares. This slow speed issue applied to download from any web browser or installing software using terminal.

    Read the article

  • Cisco PIX firewall blocking inbound Exchange email

    - by sumsaricum
    [Cisco PIX, SBS2003] I can telnet server port 25 from inside but not outside, hence all inbound email is blocked. (as an aside, inbox on iPhones do not list/update emails, but calendar works a charm) I'm inexperienced in Cisco PIX and looking for some assistance before mails start bouncing :/ interface ethernet0 auto interface ethernet1 100full nameif ethernet0 outside security0 nameif ethernet1 inside security100 hostname pixfirewall domain-name ciscopix.com fixup protocol dns maximum-length 512 fixup protocol ftp 21 fixup protocol h323 h225 1720 fixup protocol h323 ras 1718-1719 fixup protocol http 80 fixup protocol rsh 514 fixup protocol rtsp 554 fixup protocol sip 5060 fixup protocol sip udp 5060 fixup protocol skinny 2000 no fixup protocol smtp 25 fixup protocol sqlnet 1521 fixup protocol tftp 69 names name 192.168.1.10 SERVER access-list inside_outbound_nat0_acl permit ip 192.168.1.0 255.255.255.0 192.168.1.96 255.255.255.240 access-list outside_cryptomap_dyn_20 permit ip any 192.168.1.96 255.255.255.240 access-list outside_acl permit tcp any host 213.xxx.xxx.xxx eq 3389 access-list outside_acl permit tcp any interface outside eq ftp access-list outside_acl permit tcp any host 213.xxx.xxx.xxx eq https access-list outside_acl permit tcp any host 213.xxx.xxx.xxx eq www access-list outside_acl permit tcp any interface outside eq 993 access-list outside_acl permit tcp any interface outside eq imap4 access-list outside_acl permit tcp any interface outside eq 465 access-list outside_acl permit tcp any host 213.xxx.xxx.xxx eq smtp access-list outside_cryptomap_dyn_40 permit ip any 192.168.1.96 255.255.255.240 access-list COMPANYVPN_splitTunnelAcl permit ip 192.168.1.0 255.255.255.0 any access-list COMPANY_splitTunnelAcl permit ip 192.168.1.0 255.255.255.0 any access-list outside_cryptomap_dyn_60 permit ip any 192.168.1.96 255.255.255.240 access-list COMPANY_VPN_splitTunnelAcl permit ip 192.168.1.0 255.255.255.0 any access-list outside_cryptomap_dyn_80 permit ip any 192.168.1.96 255.255.255.240 pager lines 24 icmp permit host 217.157.xxx.xxx outside mtu outside 1500 mtu inside 1500 ip address outside 213.xxx.xxx.xxx 255.255.255.128 ip address inside 192.168.1.1 255.255.255.0 ip audit info action alarm ip audit attack action alarm ip local pool VPN 192.168.1.100-192.168.1.110 pdm location 0.0.0.0 255.255.255.128 outside pdm location 0.0.0.0 255.255.255.0 inside pdm location 217.yyy.yyy.yyy 255.255.255.255 outside pdm location SERVER 255.255.255.255 inside pdm logging informational 100 pdm history enable arp timeout 14400 global (outside) 1 interface nat (inside) 0 access-list inside_outbound_nat0_acl nat (inside) 1 0.0.0.0 0.0.0.0 0 0 static (inside,outside) tcp 213.xxx.xxx.xxx 3389 SERVER 3389 netmask 255.255.255.255 0 0 static (inside,outside) tcp 213.xxx.xxx.xxx smtp SERVER smtp netmask 255.255.255.255 0 0 static (inside,outside) tcp 213.xxx.xxx.xxx https SERVER https netmask 255.255.255.255 0 0 static (inside,outside) tcp 213.xxx.xxx.xxx www SERVER www netmask 255.255.255.255 0 0 static (inside,outside) tcp interface imap4 SERVER imap4 netmask 255.255.255.255 0 0 static (inside,outside) tcp interface 993 SERVER 993 netmask 255.255.255.255 0 0 static (inside,outside) tcp interface 465 SERVER 465 netmask 255.255.255.255 0 0 static (inside,outside) tcp interface ftp SERVER ftp netmask 255.255.255.255 0 0 access-group outside_acl in interface outside route outside 0.0.0.0 0.0.0.0 213.zzz.zzz.zzz timeout xlate 0:05:00 timeout conn 1:00:00 half-closed 0:10:00 udp 0:02:00 rpc 0:10:00 h225 1:00:00 timeout h323 0:05:00 mgcp 0:05:00 sip 0:30:00 sip_media 0:02:00 timeout sip-disconnect 0:02:00 sip-invite 0:03:00 timeout uauth 0:05:00 absolute aaa-server TACACS+ protocol tacacs+ aaa-server TACACS+ max-failed-attempts 3 aaa-server TACACS+ deadtime 10 aaa-server RADIUS protocol radius aaa-server RADIUS max-failed-attempts 3 aaa-server RADIUS deadtime 10 aaa-server RADIUS (inside) host SERVER *** timeout 10 aaa-server LOCAL protocol local http server enable http 217.yyy.yyy.yyy 255.255.255.255 outside http 192.168.1.0 255.255.255.0 inside no snmp-server location no snmp-server contact snmp-server community public no snmp-server enable traps floodguard enable sysopt connection permit-ipsec crypto ipsec transform-set ESP-3DES-MD5 esp-3des esp-md5-hmac crypto dynamic-map outside_dyn_map 20 match address outside_cryptomap_dyn_20 crypto dynamic-map outside_dyn_map 20 set transform-set ESP-3DES-MD5 crypto dynamic-map outside_dyn_map 40 match address outside_cryptomap_dyn_40 crypto dynamic-map outside_dyn_map 40 set transform-set ESP-3DES-MD5 crypto dynamic-map outside_dyn_map 60 match address outside_cryptomap_dyn_60 crypto dynamic-map outside_dyn_map 60 set transform-set ESP-3DES-MD5 crypto dynamic-map outside_dyn_map 80 match address outside_cryptomap_dyn_80 crypto dynamic-map outside_dyn_map 80 set transform-set ESP-3DES-MD5 crypto map outside_map 65535 ipsec-isakmp dynamic outside_dyn_map crypto map outside_map client authentication RADIUS LOCAL crypto map outside_map interface outside isakmp enable outside isakmp policy 20 authentication pre-share isakmp policy 20 encryption 3des isakmp policy 20 hash md5 isakmp policy 20 group 2 isakmp policy 20 lifetime 86400 telnet 217.yyy.yyy.yyy 255.255.255.255 outside telnet 0.0.0.0 0.0.0.0 inside telnet timeout 5 ssh 217.yyy.yyy.yyy 255.255.255.255 outside ssh 0.0.0.0 255.255.255.0 inside ssh timeout 5 management-access inside console timeout 0 dhcpd address 192.168.1.20-192.168.1.40 inside dhcpd dns SERVER 195.184.xxx.xxx dhcpd wins SERVER dhcpd lease 3600 dhcpd ping_timeout 750 dhcpd auto_config outside dhcpd enable inside : end I have Kiwi SysLog running but could use some pointers in that regard to narrow down the torrent of log messages, if that helps?!

    Read the article

  • Sending mails via Mutt and Gmail: Duplicates

    - by Chris
    I'm trying to setup mutt wiht gmail for the first time. It seems to work pretty well, however when I send a mail from Mutt i appears twice in Gmail's sent folder. (I assume it's also sent twice - I'm trying to validate that) My configuration (Stripped of coloring): # A basic .muttrc for use with Gmail # Change the following six lines to match your Gmail account details set imap_user = "XX" set smtp_url = "[email protected]@smtp.gmail.com:587/" set from = "XX" set realname = "XX" # Change the following line to a different editor you prefer. set editor = "vim" # Basic config, you can leave this as is set folder = "imaps://imap.gmail.com:993" set spoolfile = "+INBOX" set imap_check_subscribed set hostname = gmail.com set mail_check = 120 set timeout = 300 set imap_keepalive = 300 set postponed = "+[Gmail]/Drafts" set record = "+[Gmail]/Sent Mail" set header_cache=~/.mutt/cache/headers set message_cachedir=~/.mutt/cache/bodies set certificate_file=~/.mutt/certificates set move = no set include set sort = 'threads' set sort_aux = 'reverse-last-date-received' set auto_tag = yes hdr_order Date From To Cc auto_view text/html bind editor <Tab> complete-query bind editor ^T complete bind editor <space> noop # Gmail-style keyboard shortcuts macro index,pager y "<enter-command>unset trash\n <delete-message>" "Gmail archive message" macro index,pager d "<enter-command>set trash=\"imaps://imap.googlemail.com/[Gmail]/Bin\"\n <delete-message>" "Gmail delete message" macro index,pager gl "<change-folder>" macro index,pager gi "<change-folder>=INBOX<enter>" "Go to inbox" macro index,pager ga "<change-folder>=[Gmail]/All Mail<enter>" "Go to all mail" macro index,pager gs "<change-folder>=[Gmail]/Starred<enter>" "Go to starred messages" macro index,pager gd "<change-folder>=[Gmail]/Drafts<enter>" "Go to drafts" macro index,pager gt "<change-folder>=[Gmail]/Sent Mail<enter>" "Go to sent mail" #Don't prompt on exit set quit=yes ## ================= #Color definitions ## ================= set pgp_autosign

    Read the article

  • Windows Vista Wrong Certificate With SNI

    - by JamesArmes
    I'm setting up SNI on an apache server and I thought things were going well. I have two URLs from different domains that point at the same site. I have one virtual host setup for each with the appropriate certificate for each. One of the certificates is valid but the other is self-signed (waiting on GoDaddy for the real cert). If I test the different URLs in Firefox, Safari and Opera all works well. I get no errors for the URL with the valid certificate and I get a self-signed warning for the other. However, in Internet Explorer 8 and Google Chrome, both URLs return the valid certificate (even if its not valid for the specific site). So for the one site, I get a valid certificate. For the other, I get a warning about the cert being for a different site. I tried switching the order of the vhosts and it made no difference. I know that Chrome and IE both use Window's HTTP stack so I understand why the behavior is the same for the two. What I don't understand is why I'm seeing this behavior.

    Read the article

  • Windows Not Honoring DHCP Scope

    - by jerhinesmith
    Please bear with me as I'm not a networking person by trade. Our current configuration at work includes two Windows Servers serving as DHCP/Active Directory servers (if that makes sense) -- one replicating from the other. On both machines, the DNS resolution is set up as: Main Windows Box (10...* address) Public IP Address (for Verizon) Public IP Address (secondary Verizon) Secondary Windows Box (10...* address) Assuming our domain is foo.com, we maintain the foo.com website on a hosted VPS with it's own IP address. The problem is that even though bar.foo.com is an internal server and is defined in DNS on the Primary Windows machine, when I ping bar or even bar.foo.com it resolves to the hosted IP address instead of the 10.* address. I tried taking both of the Public IP addresses out of the DHCP scope, and that seemed to work, but it completely slowed down access to any external sites, so that wasn't acceptable. I also tried adding the two Windows machine as the DNS servers on my desktop. That too worked, but I'd rather not have everything enter their DNS servers, as the above setup should theoretically be working. Is there anything I could check to see why pinging bar.foo.com isn't resolving to the DNS entry on the Windows machines? Here's a summary of the ping results, if they help: Pinging from servers with static IP bar.foo.com resolves with correct IP address Pinging from linux machines not joined to the domain bar.foo.com resolves with correct IP address Pinging from user's desktop machines, joined to the domain, but dynamic IP bar.foo.com resolves with incorrect IP address This is driving me crazy!

    Read the article

  • Interaction between two Clouds

    - by Snehal Masne
    I have setup the Cloud-A with 1 - [CLC+CC] and 2 - [NC] computers. I have another Cloud-B with same configuration. [using the Ubuntu Enterprise Cloud] Both of them working fine individually, in the same LAN. Now if I want to add the NC of Cloud-A to CC of Cloud-B, [in case the resources of Cloud-B are exhausted] how can I make it possible ? I guess this calls for the interoperability stuff... Could you please explain what happens exactly when we ask for instance, the direct interaction happens between the client and NC or it goes through the CLC and CC ? What I want to say is, say there are multiple cloud providers. A user is subscribed to any one of them, say Cloud-A for IaaS. As the requirements are dynamic, all the resources of Cloud-A may get exhausted. There may be another Cloud-B which can provide the services but that Cloud-A can't ask the client to go for Cloud-B. So if it is possible to have some co-ordination between this two providers to share resources mutually, making client fully unaware of whats going on in the background....? Please reply.. I am sorry if I'm doing mistake anywhere... Thanks in advance :) Regards, www.TechProceed.com

    Read the article

  • Linux NFS create mask and force user equivalent

    - by Mike
    I have two Linux servers: fileserver Debian 5.0.3 (2.6.26-2-686) Samba version 3.4.2 apache Ubuntu 10.04 LTS (2.6.32-23-generic) Apache 2.2.14 I have a number of Samba shares on fileserver so that I can access files from Windows PCs. I am also exporting /data/www-data to the apache server, where I have it mounted as /var/www. The setup is okay, except for when I come to create files on the NFS mount. I end up with files that cannot be read by Apache, or which cannot be modified by other users on my system. With Samba, I can specify force user, force group, create mask and directory mask, and this ensures that all files are created with suitable permissions for my Apache web server. I can't find a way to do this with NFS. Is there a way to force permissions and ownership with NFS - am I missing something obvious? Although I've spent quite a bit of time with Linux, and am weaning myself off Windows, I still haven't quite got to grip with Linux permissions... If this is not the right way to do things, I am open to alternative suggestions.

    Read the article

  • On clients use generic driver for printer shared by CUPS

    - by Daniel
    I know this worked really easy with a recent CUPS version some years ago. Unfortunately I fail to do the following with latest CUPS nowadays. How to share a printer without using printer specific drivers on the clients with CUPS? The printer is a Samsung ML-2010. The important fact is that it needs a quite specific library for printing called splix. That is installed on the server and prints well. What makes trouble is using that printer over the network. I found out how to use Avahi under Linux to make use of DNSSD to advertise and discover printers. But as far as I understand the new CUPS offers the internal driver interface on the network. This has 2 major issues: anybody can fiddle with the driver and I don't trust any uncommon printer library to be "network secure" anyone who wants to use my printer including guests needs to install the specific drivers first I remember the old days when I could enable "Share this printer" on the CUPS server and all clients would magically detect the printer and just send their job data to the server and have it do the driver stuff. After everything a read I guess this is related to the changes Apple introduced with CUPS including dropping of the integrated network discovery protocol. If it helps: Server: Ubuntu LTS 12.04 Server with CUPS 1.5.3 Client: Arch Linux with current CUPS 1.6.1 On another box with Ubuntu the printer was setup automatically at least but the mechanism used the Splix library for that.

    Read the article

  • Weird Apache behaviour and with files again

    - by afifio
    Hi and thanks for stopping by. I have read Weird Apache problem with file, I have read Weird Apache problem with file ...and its not the problem Setup single XAMPP installation on Windows, single windows user, 2HD, 1 is a portable USB. All is fine, until I move the xampp to new portable HD Symptom Old php files - works fine, new one doesnt http://127.0.0.1/Ajax/index.php - yay http://127.0.0.1/test2/t.php - display the source code http://127.0.0.1/Ajax/test2/t.php - display the source code http://127.0.0.1/Ajax/t.php - display the source code Extra Info IIS+MS Web Development stuff, .NET4, Asp, etc is being installed and still hast reboot yet. .htaccess also seems doesnt work Apache2 conf file was modified to Averride All and still it doesnt care. One of the directory supposed to treat .htm as php yet got text, created another directory and edit a phpinfo, still another text, browse to phpmyadmin, viola, works fine Suspect Does Apache honour XP security and permission ? If so, this is a single user computer. Does Apache dont like my new hard disk/new place ? Why it doesnt execute the php in new directory but happily execute in old folder ? Thanks for the riddle answers

    Read the article

  • How can I use HAproxy with SSL and get X-Forwarded-For headers AND tell PHP that SSL is in use?

    - by Josh
    I have the following setup: (internet) ---> [ pfSense Box ] /-> [ Apache / PHP server ] [running HAproxy] --+--> [ Apache / PHP server ] +--> [ Apache / PHP server ] \-> [ Apache / PHP server ] For HTTP requests this works great, requests are distributed to my Apache servers just fine. For SSL requests, I had HAproxy distributing the requests using TCP load balancing, and it worked however since HAproxy didn't act as a proxy, it didn't add the X-Forwarded-For HTTP header, and the Apache / PHP servers didn't know the client's real IP address. So, I added stunnel in front of HAproxy, reading that stunnel could add the X-Forwarded-For HTTP header. However, the package which I could install into pfSense does not add this header... also, this apparently kills my ability to use KeepAlive requests, which I would really like to keep. But the biggest issue which killed that idea was that stunnel converted the HTTPS requests into plain HTTP requests, so PHP didn't know that SSL was enabled and tried to redirect to the SSL site. How can I use HAproxy to load balance across a number of SSL servers, allowing those servers to both know the client's IP address and know that SSL is in use? And if possible, how can I do it on my pfSense server? Or should I drop all this and just use nginx?

    Read the article

  • Access server using IP on another interface

    - by Markos
    I am using Windows Server 2012 instead of a router for my home network. Currently I am using RRAS and computers from local network can access Internet correctly. Here is a map of the current setup: [PC1] ---| |---- (lan ip)[Server](wan ip)--> internet [PC2] ---| I have applications running on Server, such as IIS and others. All can be accessed from internet using wan ip and from lan using lan ip. I have a domain, lets say its my-domain.com, which is resolved to my wan ip. What I want is to enable my LAN computers to be able to connect to services on my server using the very same address as internet users: eg http://my-domain.com/. However this does not work for my lan computers. What I understand is that I need to set up some kind of loopback route in a way that packets comming to LAN interface get routed to WAN interface. But I haven't found how to achieve this (in fact, I don't know WHAT to search for). Feel free to ask for additional informations and I will try to update the question.

    Read the article

  • Nginx + WordPress + HHVM: Why isn't Batcache working? Would Varnish help even more?

    - by javipas
    I've heard great things about HHVM, so I've setup a copy of WordPress blog (on another domain) with Nginx (with the Pagespeed module) and HHVM. Right now the benefits are obvious: on the same config, load times are between two and three times faster. I'm trying to speed up things a little bit, and I've also installed Memcached and Batcache. I've installed the memcached package, copied object-cache.php (Pastebin) onto the root folder of the WordPress blog, and after that I've installed the Batcache plugin and copied the advanced-cache.php (Pastebin) file onto the wp-content folder. Also, I've included the line define('WP_CACHE', true); in the wp-config.php file. It seems it doesn't work, though. If I quickly reload the page several times Batcache should show the cached page, but it doesn't. It's easy to check that by reloading (Cmd+R on Chrome on OS X) the page several times and then viewing the page's code. Under the <head> section I should see some batcache stats, but they aren't there. I wonder if someone could give me some hint on this. On a side note, I don't know if I could add some other component in order to help the performance be even better. I'm thing about Varnish, but I'm not sure if it's just useless and it's just another way to the same I'm currently doing. Any other component there? (I'll test CDN for images, minifying js, etc and some other tricks as well, but I'm talking from the server perspective).

    Read the article

  • xen + debian network after upgrade squeeze to wheeze

    - by rush
    I've got a Debian + Xen server. After a system upgrade to the stable version the network doesn't come up after boot. Every time after reboot I need to bring it up manually. The network configuration was not changed during upgrade. Here is /etc/network/interfaces: auto lo iface lo inet loopback auto eth0 iface eth0 inet static address 11.22.33.44 netmask 255.255.255.0 gateway 11.22.33.1 nameserver 8.8.8.8 After boot ip r shows no route and eth0 has no ip address. Manually ip and route setup goes fine and network starts working. Messages from dmesg about network I've found (looks like nothing interesting) [ 3.894401] ACPI: Fan [FAN3] (off) [ 3.894444] ACPI: Fan [FAN4] (off) [ 4.178348] e1000e 0000:00:19.0: eth0: (PCI Express:2.5GT/s:Width x1) 00:1e:67:14:66:c9 [ 4.178351] e1000e 0000:00:19.0: eth0: Intel(R) PRO/1000 Network Connection [ 4.178392] e1000e 0000:00:19.0: eth0: MAC: 10, PHY: 11, PBA No: 0100FF-0FF [ 4.178413] e1000e 0000:02:00.0: Disabling ASPM L0s L1 [ 4.178432] xen: registering gsi 16 triggering 0 polarity 1 -- [ 4.223667] ata5: DUMMY [ 4.223668] ata6: DUMMY [ 4.289153] e1000e 0000:02:00.0: eth1: (PCI Express:2.5GT/s:Width x1) 00:1e:67:14:66:c8 [ 4.289155] e1000e 0000:02:00.0: eth1: Intel(R) PRO/1000 Network Connection [ 4.289245] e1000e 0000:02:00.0: eth1: MAC: 3, PHY: 8, PBA No: 1000FF-0FF [ 4.506908] usb 1-1: new high-speed USB device number 2 using ehci_hcd [ 4.542920] ata2: SATA link up 3.0 Gbps (SStatus 123 SControl 300) -- [ 10.362999] EXT4-fs (dm-23): mounted filesystem with ordered data mode. Opts: (null) [ 10.419103] EXT4-fs (dm-3): mounted filesystem with ordered data mode. Opts: (null) [ 10.988255] ADDRCONF(NETDEV_UP): eth1: link is not ready [ 13.175533] Event-channel device installed. [ 13.287555] XENBUS: Unable to read cpu state -- [ 13.288670] XENBUS: Unable to read cpu state [ 13.965939] Bridge firewalling registered [ 14.134048] e1000e: eth1 NIC Link is Up 1000 Mbps Full Duplex, Flow Control: Rx/Tx [ 14.283862] ADDRCONF(NETDEV_UP): peth0: link is not ready [ 14.284543] ADDRCONF(NETDEV_CHANGE): eth1: link becomes ready [ 17.800627] e1000e: peth0 NIC Link is Up 1000 Mbps Full Duplex, Flow Control: Rx/Tx [ 17.801377] ADDRCONF(NETDEV_CHANGE): peth0: link becomes ready [ 18.307278] device peth0 entered promiscuous mode [ 24.538899] eth1: no IPv6 routers present [ 28.570902] peth0: no IPv6 routers present I've upgraded two servers and I've such behaviour on two of them. How to fix this and get network starts automatically on boot?

    Read the article

  • SSD for swap on Ubuntu server

    - by grs
    Currently I am reading SSD reviews and I wonder how much exactly I will benefit if I move the 24 GB swap from 7200rpm HDD to SSD. Does anyone implemented swap space on SSD? Is this generally good idea? On a side note: I read that ext4 has much better performance if the journal is on SSD. Anyone with such a setup? Thanks! Edit: Here I will answer the questions posted: Occasionally, relatively rare I am hitting the swap. I know what the swap is for and that is better to get more RAM. When the server begins to swap its performance degrades (not a surprise). The idea is if I have few memory hungry processes running, to improve the overall system performance at that time, using SSD for swap, instead of slower rotational media. At the end - I want to be able to login faster and check the server state during swapping, instead of waiting on the login prompt. And of what I see SSD is cheaper per GB than RAM. Would I have better server performance during swapping (as rare it is) using SSD compared to HDD? Where 10k or 15k rpm HDDs would rate in this scenario? Thank you all for your quick and prompt answers!

    Read the article

  • How can I configure Samba to share (read/write) any folder with root permissions?

    - by Mike Toews
    I have a CentOS 5 VirtualBox guest on a Win7x64 host. I am attempting to setup a read/write share a directory owned by root with my Windows host using Samba, but I'm having no luck after running around in circles. To simplify matters, I've disabled my Firewall (/etc/init.d/iptables stop). As security and permissions are irrelevant for this purpose, I'd rather not have to set up another unix user/group/password. Here is the output from testparm Load smb config files from /etc/samba/smb.conf rlimit_max: rlimit_max (1024) below minimum Windows limit (16384) Processing section "[Guest Share]" Loaded services file OK. Server role: ROLE_STANDALONE and the source of /etc/samba/smb.conf: [global] workgroup = WRKGRP netbios name = SMBSERVER security = SHARE load printers = No [Guest Share] comment = Guest access share path = /root/src read only = No guest ok = Yes Running /etc/init.d/smb restart shows an OK status. However, on my Windows host, I can only see the share folder on the guest \\IPv4, but I cannot go into "Guest Share": "The network name cannot be found" error message is a common error, with a likely cause: The user you are trying to access the share with does not have sufficient permissions to access the path for the share. Both read (r) and access (x) should be possible. Am I trying to use root as a passwordless Samba guest? I'd like to, is it possible? How can I configure Samba to share (read/write) any folder with root permissions?

    Read the article

  • vSphere Promiscuous mode only receiving packets one way from network switch

    - by steve.lippert
    We have two network switches, a POE switch (SwitchA) to power our phones / users computers and a non-POE switch (SwitchB for the rest network.) Each switch is setup to do port mirroring to support our VoIP recording system. SwitchA does port mirroring on specific ports if we need to record a user. SwitchB mirrors one port to monitor our work at home users (Internet comes in from managed router, to switch, back out to our firewall.) These two port mirroring setups feed into one vmware vSphere 4.1 server, it has four total physical cards. The other two NICs feed into an unmanaged switch for connecting to the rest of the network. Once into the vSphere server all network ports go into a vSwitch, and then one of the servers (Windows 2008 R2) sniffs them out and does its thing. Everything is working fine and dandy from SwitchB. But on SwitchA we only receive one side of the VoIP packets (going out to the phone, nothing coming in from the phone). Troubleshooting steps I have taken so far: I hooked up my laptop to the monitor port on SwitchB and I see both sides of the packets. I swapped which network interface is plugged into the monitor port on SwitchA. Because everything feeds into one vSwitch / vNetwork and both sides of the conversation arrive just fine from SwitchB I believe everything is configured correctly on the vSphere server/guest. What could be causing one way packets to arrive on my guest machine from only one interface, but not the other? Could a bad cable be causing the problems from SwitchB?

    Read the article

  • Advertise a subnet route with radvd

    - by Thomas Berger
    we have set up a small IPv6 Testing network. The setup looks like this: ::/0 +----------+ | Firewall | Router to the public net +----------+ | 2001:...::/106 | +----------+ +-------| SIT GW | sit Tunnel gatway to the some test users | +----------+ | +----------+ | Test Sys | Testsystem +----------+ The idea is to advertise the default route from the firewall and the route for the SIT subnets from the sit gateway. The configurations for radvd are: # Firewall interface eth0 { AdvSendAdvert on; route ::/0 { }; }; # SIT Gatway interface eth0 { AdvSendAdvert on; route 2001:...::/106 { }; }; We have captured the adv. packages with tcpdump and the packages looks good. We see a default route from the fw, and the subnet route from the SIT gatway. But if we look on the testsystem there are two default routes over both gateways. There is no subnet route. The routing does not work of course. Here the routes we get: 2001:.....::/64 dev eth0 proto kernel metric 256 mtu 1500 advmss 1440 hoplimit 4294967295 fe80::/64 dev eth0 proto kernel metric 256 mtu 1500 advmss 1440 hoplimit 4294967295 default via fe80::baac:6fff:fe8e:XXXX dev eth0 proto kernel metric 1024 expires 0sec mtu 1500 advmss 1440 hoplimit 64 default via fe80::e415:aeff:fe12:XXXX dev eth0 proto kernel metric 1024 expires 0sec mtu 1500 advmss 1440 hoplimit 64 Any Idea?

    Read the article

  • How do I repartition an SDHC card in Windows?

    - by Peter Mortensen
    How do I repartition an SDHC card (4 GB or more)? Do I need third-part tools or Linux (a live CD solution would be OK)? In Windows' Disk Management the option Delete Partition is dimmed out: I can reformat the card as FAT32, copy files to and from the card and even change the file system to NTFS using the command line command CONVERT, but not repartition it. The article How to Partition an SD Card in Windows XP talks about using "a Windows enabler program" which sound rather dubious to me. I have tried to change from “Optimize for quick removal” to “Optimize for performance”. The option to format as NTFS appeared, but the Delete Partition option is still dimmed out. Platform: Windows XP 64-bit SD card reader: USB 2.0 device, LogiLink® CR0005C Cardreader 3,5' USB 2.0 intern 54-in-1 mit USB Front Kingston 16 GB SDHC card, speed class 4. (It could be formatted as FAT32 and successfully used in a 4 GB ReadyBoost setup (Windows 7).) I have also tried on different versions of Windows and with different cards with the same result: Kingston 4 GB SDHC card, speed class 4 (the one shown in the screenshot) Transcend 2 GB (not marked as SDHC, but SD) Windows 7 32-bit (albeit with a somewhat an older card reader) and Windows XP 32-bit on an EliteBook 8730w

    Read the article

< Previous Page | 563 564 565 566 567 568 569 570 571 572 573 574  | Next Page >