Search Results

Search found 27229 results on 1090 pages for 'default aspx'.

Page 873/1090 | < Previous Page | 869 870 871 872 873 874 875 876 877 878 879 880  | Next Page >

  • pam_tally2 causing unwanted lockouts with SCOM or Nervecenter

    - by Chris
    We use pam_tally2 in our system-auth config file which works fine for users. With services such as SCOM or Nervecenter it causes lockouts. Same behavior on RHEL5 and RHEL6 This is /etc/pam.d/nervecenter #%PAM-1.0 # Sample NerveCenter/RHEL6 PAM configuration # This PAM registration file avoids use of the deprecated pam_stack.so module. auth include system-auth account required pam_nologin.so account include system-auth and this is /etc/pam.d/system-auth auth sufficient pam_centrifydc.so auth requisite pam_centrifydc.so deny account sufficient pam_centrifydc.so account requisite pam_centrifydc.so deny session required pam_centrifydc.so homedir password sufficient pam_centrifydc.so try_first_pass password requisite pam_centrifydc.so deny auth required pam_tally2.so deny=6 onerr=fail auth required pam_env.so auth sufficient pam_unix.so nullok try_first_pass auth requisite pam_succeed_if.so uid >= 500 quiet auth required pam_deny.so account required pam_unix.so account sufficient pam_succeed_if.so uid < 500 quiet account required pam_permit.so password requisite pam_cracklib.so try_first_pass retry=3 minclass=3 minlen=8 lcredit=1 ucredit=1 dcredit=1 ocredit=1 difok=1 password sufficient pam_unix.so sha512 shadow try_first_pass use_authtok remember=8 password required pam_deny.so session optional pam_keyinit.so revoke session required pam_limits.so session [success=1 default=ignore] pam_succeed_if.so service in crond quiet use_uid session required pam_unix.so The login does work but it also triggers the pam_tally counter up until it hits 6 "false" logins. Is there any pam-ninjas around that could spot the issue? Thanks.

    Read the article

  • Port Forwarding to put my web server on The Internet

    - by Chadworthington
    I went to http://canyouseeme.org/ to check to see what my external IP address. Regardless of what port I enter, it tells me that the port is blocked. I have a LinkSys router that basically has the default settings with the exception that I have WEP encrptin setup and I have forwarded a few ports, including 80 and 69. I forwarded them to the 192.x.x.103 IP address of the PC which is running IIS. That PC runs Symantec Endpoint Protection, which I right mouse clicked in the tray to Disable. These steps used to make my PC visible so I could host my own web site in IIS on port 80, or some other port, like 69. Yet, the Open Port tool cannot see my IP when it checks eiether port and when I navigate to http://my external ip/ I get "page cant be displayed" At first I was thinking that maybe Comcast is blocking port 80, but 69 doesnt work eiether. I do not see any other blockking set up in my router and, as I mentioned, I went with teh defaults except where discussed. This is a corporate PC and Symantec End Point Protecion is new to it (this previously worked on teh same PC with Symantec Protection Agent), but I thought that disabling Sym End Pt from the tray, that that would effectively neutralize it. I do not have the rights to kill the program itself. Any suggestions on what else to try to make my PC externally visible?

    Read the article

  • How to completely disable IPv6 for loopback interface on RHEL 5.6

    - by Marc D
    I've done lots of research on how to disable IPv6 on RedHat Linux and I have it almost completely disabled. However the loopback interface is still getting an inet6 loopback address (::1/128). I can't find where IPV6 is still enabled for loopback. To disable IPV6 I added the following settings to /etc/sysctl.conf: net.ipv6.conf.default.disable_ipv6=1 net.ipv6.conf.all.disable_ipv6=1 And also added the following line to /etc/sysconfig/network: NETWORKING_IPV6=no After rebooting, the inet6 address is gone from my physical interface (eth0), but is still there for lo: # ip addr show 1: lo: <LOOPBACK,UP,LOWER_UP> mtu 16436 qdisc noqueue link/loopback 00:00:00:00:00:00 brd 00:00:00:00:00:00 inet 127.0.0.1/8 scope host lo inet6 ::1/128 scope host valid_lft forever preferred_lft forever 2: eth0: <BROADCAST,MULTICAST,UP,LOWER_UP> mtu 1500 qdisc pfifo_fast qlen 1000 link/ether 00:50:56:xx:xx:xx brd ff:ff:ff:ff:ff:ff inet 10.x.x.x/21 brd 10.x.x.x scope global eth0 If I manually remove the IPV6 address from loopback and then bounce the interface, it comes back: # ip addr del ::1/128 dev lo # ip addr show lo 1: lo: <LOOPBACK,UP,LOWER_UP> mtu 16436 qdisc noqueue link/loopback 00:00:00:00:00:00 brd 00:00:00:00:00:00 inet 127.0.0.1/8 scope host lo # ip link set lo down # ip link set lo up # ip addr show lo 1: lo: <LOOPBACK,UP,LOWER_UP> mtu 16436 qdisc noqueue link/loopback 00:00:00:00:00:00 brd 00:00:00:00:00:00 inet 127.0.0.1/8 scope host lo inet6 ::1/128 scope host valid_lft forever preferred_lft forever I believe IPV6 should be disabled at the kernel level as confirmed by sysctl: # sysctl net.ipv6.conf.lo.disable_ipv6 net.ipv6.conf.lo.disable_ipv6 = 1 Any ideas on what else would cause the loopback interface to get an IPV6 address?

    Read the article

  • No sound out of headphone port on laptop

    - by Thanatos
    I cannot get sound out of the headphone port on a laptop. Headphones are plugged in, and sound comes out of the internal speakers. Windows behaves normally (sound switches to headphones when headphones are inserted). It did work in Linux at one point, but something changed, we're just not sure what. Rebooting doesn't fix. This appears to occur whether or not PulseAudio is running. Things I've tried: Rebooting. No effect. Booting into Windows. It works properly, so probably not a hardware issue. All of alsamixer. My only controls are this: "Master" Volume bar & mutable, unmuted. Controls volume. "PCM" Volume bar only. 100%. "S/PDIF" Mutable only, currently muted, has no effect. "S/PDIF" Default PCM", Mutable only, currently unmuted, has no effect. Killing PulseAudio. No effect. (It also won't stay dead! Something appears to be restarting it, and I can't tell what, but it is annoying as fuck.) alsactl init 0, no effect. sudo rm -f /var/lib/alsa/asound.state, no effect. General system info: Ubuntu 10.04 LTS Toshiba Satellite T135D-S1324 lspci says I have: 00:14.2 Audio device: ATI Technologies Inc SBx00 Azalia (Intel HDA) 01:05.1 Audio device: ATI Technologies Inc RS780 Azalia controller Some edits: Yes, the headphones are in all the way. This works in Windows: You plug headphones in, the internal speakers stop making noise, and noise comes out the head phones. Windows says I only have two sound cards: the HDMI port (which I don't care about) and the "sound card", which it claims is a "Conexant Pebble High Definition SmartAudio" In Windows, both the internal speakers and the headphone jack show up as one soundcard, which in my experience, is typical. (This is a laptop)

    Read the article

  • How to route broadcast packets from machine with two network interfaces on same subnet

    - by Syam
    I run RHEL 5 and have two NICs on one machine connected to the same subnet: eth0 192.168.100.10 eth1 192.168.100.11 My application needs to receive and transmit UDP packets (both unicast & broadcast) via these interfaces. I've found the way to handle the ARP problem and I've added routes to handle the routing problem: ip rule add from 192.168.100.10 lookup 10 ip route add table 10 default src 192.168.100.10 dev eth0 (and similarly, table 11 for eth1) The problem is that only unicast packets gets routed properly. Broadcast packets always go out through eth0. I tried removing the rule for 192.168.100.0 & 192.168.100.255 from table 255 and adding them to my tables. But then I see ARP requests being given out for packets to 192.168.100.255 (obviously, no nodes respond and nobody gets any data). Due to several techno-political issues, I'm stuck with this configuration and can't change subnets or try something different. I've tried SO_BINDTODEVICE and it works, but I'd prefer a solution that doesn't need my application to run as root. Is there a way to get this working? Any help is highly appreciated.

    Read the article

  • Intermittently uncommunicative subnets

    - by mhd
    Last week proved me a veritable Cassandra: I've always said that it's a bad idea to have only one firewall/router, without a backup or failover. And thus our Cisco PIX went haywire, refusing to route properly. And of course, the only one available here on short notice is me, and while I'm quite grounded in Linux, I'm really a developer not a sysadmin (the fact that this hit me on sysadmin appreciation day is a bit ironic). Anyway, this weekend I tried to hack up a temporary solution: I used an old server with enough NICs (two built-in, four on a card) to serve as a gateway and firewall. Due to some problems with the raid controller, I got only two router distros running, and between Untangle and Ebox I decided for the latter. Now everything is quite okay. I've got all the different subnets we've got here (all with separate switches) talking to each other and even to the internet (Cisco 2800 router, T1 lines). But from time to time (20-60 minute intervals), I get a total routing failure. Our main, office subnet can't talk to our server subnet and can't connect to the internet. This is not the end of a gradual slowdown, either everything's working perfectly or I get a total lack of communication for about two minutes each time. Now I'm a bit at wits end what to check. At least with the default EBox setup, nothing in /var/log shows anything weird and it doesn't exactly have lots of built-in monitoring tools. So I'm hoping someone here could give me some pointers about what to look out for. I did change the ethernet cable from the office switch to the firewall, with no results. I might change switches, although within the switch it seems to work ok enough. Edit: I'm not sure whether this is the sole cause of the problem, but after I noticed a few DHCP entries just before the last drop of connectivity, I tried to reproduce that. And alas, whenever I renew a DHCP connection, I can't access other subnets anymore. Running ISC DHCPD 3.0.6.

    Read the article

  • Server 2003 and XP Client; Why are HTTP connections being silently dropped.

    - by Asa Yeamans
    On my network, my edge-router, a windows 2003 r2 server router with all the latest updates, will drop packets, but only under specific circumstances. I have troubleshot and isolated it down to the most simple configuration i can. There is NO NAT involved. Only fully-public IP addresses. No Firewalls are running either, all ahve been disabled. no packet filters on any interfaces anywhere either. I have a single Windows XP virtual machine and my edge-router(the windows 2003 r2 server, and also a virtual machine) running on a windows 2008 x64 r2 system (running virtual server 2005 as i dont have Intel-VT compatible chip yet). The edge router can access any external http site just fine, no issues. However the windows XP machine is only able to access certain sites. These work: www.google.com www.txstate.edu www.workintexas.com www.thedailywtf.com . These Dont: www.yahoo.com www.utexas.edu en.wikipedia.org slashdot.org www.bing.com. I have removed all possibility of DNS issues by connecting with net-cat from the XP box and sending GET /\r\nHost: \r\n\r\n and that connection replicates the issue as well. The network setup: My statically assigned IP block: x.x.x.168/29 DSL Modem -----PPPoE Connection---- x.x.x.169[EdgeRouter] [EdgeRouter]x.x.x.170 -----Virtual Ethernet----- x.x.x.174 [Test2] Test2's Default gateway is x.x.x.170 and test2 can ping any and every valid, accessible, public IP address with no packet loss what-so-ever. If i connect directly over PPPoE from test2 (the XP box) everything works just fine... Im at my wits end, i have NO IDEA whats causing this.

    Read the article

  • Wired to wireless bridge in Linux

    - by adrianmcmenamin
    I am attempting to set up my Raspberry Pi as a bridge (but I think this is not a question specific to the hardware) - using Debian wheezy. I have a hostapd.conf: (some details changed for security)... interface=wlan0 bridge=br0 driver=nl80211 auth_algs=1 macaddr_acl=0 ignore_broadcast_ssid=0 logger_syslog=-1 logger_syslog_level=0 hw_mode=g ssid=MY_SSID channel=11 wep_default_key=0 wep_key0=MY_KEY wpa=0 (yes, I know WEP is no good) And this in /etc/network/interfaces auto lo iface lo inet loopback iface eth0 inet dhcp allow-hotplug wlan0 iface wlan0 inet manual wpa-roam /etc/wpa_supplicant/wpa_supplicant.conf iface default inet dhcp auto br0 iface br0 inet dhcp bridge-ports eth0 wlan0 Everything seems to come up ok, but I cannot associate with the bridged wireless connection - even though the flashing lights on the USB stick suggest packets are being exchanged. I have read somewhere that not all cards/devices will run in hostap mode - they won't pass packets in one direction: is that right? (The info was a bit old)- this my card: [ 3.663245] usb 1-1.3.1: new high-speed USB device number 5 using dwc_otg [ 3.794187] usb 1-1.3.1: New USB device found, idVendor=0cf3, idProduct=9271 [ 3.804321] usb 1-1.3.1: New USB device strings: Mfr=16, Product=32, SerialNumber=48 [ 3.816994] usb 1-1.3.1: Product: USB2.0 WLAN [ 3.823790] usb 1-1.3.1: Manufacturer: ATHEROS [ 3.830645] usb 1-1.3.1: SerialNumber: 12345 So, what have I got wrong here?

    Read the article

  • Defining Virtual and Real User Directories with Dovecot & Postfix

    - by blankabout
    Following a wobble described in this question we now have virtual and real users authenticating with Dovecot, the problem now is that the real users (who have been on the system for years) can no longer access their mail. I'm guessing that it is because Dovecot is configured to point to the virtual mailboxes but not the real mail boxes. These are snippets from the config files: /etc/dovecot/dovecot.conf !include conf.d/*.conf /etc/dovecot/conf.d/10-auth.cong passdb { driver = passwd-file # Path for passwd-file. Also set the default password scheme. args = scheme=cram-md5 /etc/cram-md5.pwd } userdb { driver=static #args = mail_uid=dovecot mail_gid=dovecot /etc/dovecot/userdb args = uid=vmail gid=vmail home=/var/spool/vhosts/%d/%n /etc/dovecot/userdb } [email protected]:::::/var/spool/vhosts/virtualdomain.com/:/bin/false:: We think the problem is that the Dovecot file 10-auth.conf does not contain a method of accessing the mailboxes for the real users. We have looked around on this site, dovecot.org and done the usual googling but cannot find anywhere that describes how to set up virtual users on alongside legacy real users. Any help would be appreciated, especially by our real users who would like the contents of their inboxes to be available! If any further config is required, please let me know.

    Read the article

  • VPN into multiple LAN Subnets

    - by Rain
    I need to figure out a way to allow access to two LAN subnets on a SonicWall NSA 220 through the built-in SonicWall GlobalVPN server. I've Googled and tried everything I can think of, but nothing has worked. The SonicWall NSA management web interface is also very unorganized; I'm probably missing something simple/obvious. There are two networks, called Network A and Network B for simplicity, with two different subnets. A SonicWall NSA 220 is the router/firewall/DHCP Server for Network A, which is plugged into the X2 port. Some other router is the router/firewall/DHCP server for Network B. Both of these networks need to be managed through a VPN connection. I setup the X3 interface on the SonicWall to have a static IP in the Network B subnet and plugged it in. Network A and Network B should not be able to access each other, which appears the be the default configuration. I then configured and enabled VPN. The SonicWall currently has the X1 interface setup with a subnet of 192.168.1.0/24 with a DHCP Server enabled, although it is not plugged in. When I VPN into the SonicWall, I get an IP address supplied by the DHCP Server on the X1 interface and I can access Network A remotely although I do not have access to Network B. How can I allow access to both Network A and Network B to VPN clients although keep devices on Network B from accessing Network A and vice-versa. Is there some way to create a VPN-only subnet (something like 10.100.0.0/24) on the SonicWall that can access Network A and Network B without changing the current network configuration or allowing devices on both netorks "see" each other? How would I go about setting this up? Diagram of the network: (Hopefully this kind of helps) WAN1 WAN2 | | [ SonicWall NSA 220 ]-(X3)-----------------[ Router 2 ] | | (X2) 192.168.2.0/24 10.1.1.0/24 Any help would be greatly appriciated!

    Read the article

  • Disabling LDAP Signing on Windows PDC in Local Policy

    - by Golmaal
    I just tripped over my own feet it seems. Playing around on a Windows 2008 R2 server (set up as domain controller), I was intrigued by certain warning event (event id 2886) which says: "To enhance the security of directory servers, you can configure both Active Directory Domain Services (AD DS) and Active Directory Lightweight Directory Services (AD LDS) to require signed Lightweight Directory Access Protocol (LDAP) binds." So I thoughtlessly did some Googling and set the relevant policies which enforce LDAP signing. Now I don't remember but I may have done that using Local Policy. Now I have setup a pfsense box which must authenticate AD users via LDAP. While the firewall can communicate over secure channel, it is difficult to manage the same for other packages such as Squid and SquidGuard. So now I have to disable i.e. undo those policy changes. The problem is that they are greyed out! The policies in question are LDAP server signing and LDAP client signing. I don't remember what I did but when I access these policies from Local Policy editor on the server, they are set to "Require Signing" and are greyed out. The same policies can still be set via Default Domain Controller option in Group Policy editor. So how can I reset these greyed out policies? Thanks

    Read the article

  • Collect temperature and fan speed with munin from Windows 7 PC?

    - by nfm
    Hi, I'm quite fond of munin and using it also at home to monitor my PCs. What was super-duper easy under Linux is pretty much unsolvable for me under Windows: I'd like to monitor CPU and Motherboard temperatures as well as fan speed. On Linux I'm using lm-sensors and the plugin for munin was basically there. I access already some information from my Windows machine via SNMP (disk space, CPU usage, memory usage); the graphs are simple as is the information exposed via SNMP, but they do their job. But when it comes to temperature and fan speed I'm running against a wall. My research so far resulted in that Windows does not by default provide out of the box ability to retrieve temperature/fan speed data. Third party applications are necessary which have know-how how to communicate with the Motherboard chips. The best I cam up with is that SpeedFan exposes a shared memory interface and there exists a library which hooks into Windows SNMP facility and bridges over to SpeedFans shared memory interface; it's called SFSNMP (site currently down). Unfortunately the library doesn't work, there's a bug report at SpeedFan open about it, but it's currently not moving (although the SFSNMP author is active there) . So, unless that's going to work like anytime soon, are there any alternatives? I'm not found of buying any software to get that feature, given that I take it as granted that my system exposes me the information to properly monitor it, but anyway don't just not answer because of this.

    Read the article

  • Adding Multiple Interfaces to EC2 Ubuntu 12.04

    - by nocode
    I have a m1.medium Ubuntu 12.04 instance with two ENI's. I have a VPC setup with a private and public subnet. Private: 10.50.1.0/24 Public: 10.50.101.0/24 I initiated the instance on the private subnet. I configured a NAT instance and route all servers in the private subnet internet access. The route tables on the private subnet point towards the NAT instance and the route table on the public subnet point to the internet gateway. I am trying to add a public interface on the machine so that I can put it behind a ELB. When I added the second ENI and configured a static IP in /etc/network/interfaces and restarted the network services, I can no longer access from the Public subnet to the Private Subnet. Works Private private Private public Does not work Public private From Public Private, I ran a TCPDUMp on the private machine and can see the request coming in. My guess is it's trying to route over the new Public interface instead of the Private. Here's my route: default 10.50.1.1 0.0.0.0 UG 100 0 0 eth0 10.50.1.0 * 255.255.255.0 U 0 0 0 eth0 10.50.101.0 * 255.255.255.0 U 0 0 0 eth1 My networking knowledge is limited and I believe I have to add some routes but unsure of what command/syntax needs to be.

    Read the article

  • Problems Installing slapd On Ubuntu Server 11.10

    - by Zach Dziura
    I know that there's a Ubuntu-specific StackExchange website, but I thought that I'd ask here because it's a server-specific question. If I'm wrong in my logic... Well, you people are better at this than I am! O=) On with the show! I'm in the process of installing Oracle Database 11g R2 Standard Edition onto Ubuntu Server 11.10. I found a guide on the Oracle Support Forums that walks you through the process fairly easily. Unfortunately, I'm running into issues installing one particular dependency: slapd. When I go to install it, I get this error message: (Reading database ... 64726 files and directories currently installed.) Unpacking slapd (from .../slapd_2.4.25-1.1ubuntu4.1_amd64.deb) ... Processing triggers for man-db ... Processing triggers for ufw ... Processing triggers for ureadahead ... Setting up slapd (2.4.25-1.1ubuntu4.1) ... Usage: slappasswd [options] -c format crypt(3) salt format -g generate random password -h hash password scheme -n omit trailing newline -s secret new password -u generate RFC2307 values (default) -v increase verbosity -T file read file for new password Creating initial configuration... Loading the initial configuration from the ldif file () failed with the following error while running slapadd: str2entry: invalid value for attributeType olcRootPW #0 (syntax 1.3.6.1.4.1.1466.115.121.1.15) slapadd: could not parse entry (line=1051) dpkg: error processing slapd (--configure): subprocess installed post-installation script returned error exit status 1 Errors were encountered while processing: slapd E: Sub-process /usr/bin/dpkg returned an error code (1) After much Google searches and forum trolling, I have yet to find a definitive answer as to what's going wrong. The error messages seem straight forward enough, but I have no idea how to debug this. Can anyone offer some assistance? Again, if I'm asking in the wrong place, I apologize. If I'm indeed asking properly, then thank you for any and all help!

    Read the article

  • Why did I loose access to the mailboxes on my old web/mail host after changing to a new one but keeping old MX values

    - by LaserBeak
    So I changed the NS records with registrar to point at the new webhosts DNS servers and edited the SOA record there, deleting the new hosts default MX records and instead putting in the old ones for the old web\mail hosts. The website A record is however pointing at the new webhosts servers and the site comes up fine. But none of this should cause me to loose access to mailboxes on my old hosts mail server right? I log into the control panel on the old host, all the mailboxes are there, all the passwords are fine but I can't log in using either webmail or pop3, says incorrect log-in/password. I even created a new mailbox and password for it respectively, but it would not let me log in. For what its worth I did not change\delete the records for 'A' on the old webhost zone file, since I am not hosting the site with them anymore and NS records are pointing to other hosts DNS servers/zone file so that shouldn't matter right? The old hosts mailserver is also not simply down, I can tell because through the control panel I setup a mail forward for one of the existing inboxes and when sending mail to it, it receives it and forwards it fine. So from this I can deduce that I have correctly inputted the old hosts MX records into the zone file hosted on the new hosts DNS and the mail is being sent to the old hosts mail server(s) and is successfully forwarded by it. But why can't I log into those account/inboxes anymore ?

    Read the article

  • SSLVerifyClient optional with location-based exceptions

    - by Ian Dunn
    I have a site that requires authentication in order to access certain directories, but not others. (The "directories" are really just rewrite rules that all pass through /index.php) In order to authenticate, the user can either login with a standard username/password, or submit a client-side x509 certificate. So, Apache's vhost conf looks something like this: SSLCACertificateFile /etc/pki/CA/certs/redacted-ca.crt SSLOptions +ExportCertData +StdEnvVars SSLVerifyClient none SSLVerifyDepth 1 <LocationMatch "/(foo-one|foo-two|foo-three)"> SSLVerifyClient optional </LocationMatch> That works fine, but then large file uploads fail because of the behavior documented in bug 12355. The workaround for that is to set SSLVerifyClient require (or optional) as the default, so now the conf looks like this SSLCACertificateFile /etc/pki/CA/certs/redacted-ca.crt SSLOptions +ExportCertData +StdEnvVars SSLVerifyClient optional SSLVerifyDepth 1 <LocationMatch "/(bar-one|bar-two|bar-three)"> SSLVerifyClient none </LocationMatch> That fixes the upload problem, but the SSLVerifyClient none doesn't work for bar-one, bar-two, etc. Those directories are still prompted to present a certificate. Additionally, I also need the root URL to accessible without the user being prompted for a certificate. I'm afraid that will cancel out the workaround, though.

    Read the article

  • Winamp has slow /skipping video playback on Windows 7

    - by Roy Rico
    Hello, I have Windows 7 x64 (7600 90-day trial version) and Winamp 5.6 installed. When I play a video in Windows Media Player, the video plays smooth, however when I play a video in winamp, the video is mostly ok when played back at the original size (but not completely), but if I play it back in fullscreen, the playback gets really slow. The video's audio track plays just fine. I have a DELL XPS 420 computer (8GB of RAM) with a Nvidia GeForce 8800 CTS 512 video card. I've updated to the latest drivers. I have the default Windows 7 codecs, and the CCCP codec pack which used to be all I needed under Windows XP to play all types of videos. Are the codecs needed for Windows Y the same? What's going on? UPDATE: As suggested, I turned off Aero and winamp ran just fine again. So I just have to wait for winamp to be rewritten to work with the way Vista/Windows 7 runs? UPDATE 2: Winamp has updated their player, and it works great with Windows 7 now.

    Read the article

  • IIS not responding with SSL Server Hello

    - by Damien_The_Unbeliever
    I'm having difficulty getting a particular IIS machine to "do" SSL. This is a test environment (one of many) which we've set up "the same" as we have many times previously, but it just doesn't seem to be working. The server is Windows Server 2003 (Version 5.2 (Build 3790.srv03_sp2_gdr.100216-1301 : Service Pack 2)) IIS is hosting 4 sites (including the default site), but only one site is configured for SSL. We're using the same SSL certificate we use on all of our other servers (it's a wildcard cert). (Don't think this is relevant, but including anyway) We've configured the site to require SSL, it has a subdirectory that doesn't require SSL but has an asp page that redirects into SSL. The 403;4 error page for the site is hooked up to this asp page (this is how we do our non-HTTPS into HTTPS transition). Using Microsoft Network Monitor (3.3), I've just run a session against a server where SSL is working. It can pull apart the SSL exchange as the following messages: SSL: Client Hello SSL: Server Hello. Certificate. Server Hello Done SSL: Client Key Exchange. Change Cipher Spec. Encrypted Handshake Message. SSL: Change Cipher Spec. Encrypted Handshake Message However, on our problem server, I only see: SSL: Client Hello. The next packet from the server (from port 443, so it's listening and responding there) contains only 60 bytes, and just seems to have the Tcp headers and not much else (but I'm by no means an expert at deciphering these things). So, where do I look next? Or what additional information do I need to add to this question, and where do I find it?

    Read the article

  • ubuntu preseed installation keep missing mirror files

    - by JackWu
    Install ubuntu12.04.2 with preseed file, but there is one buggy problem about preseed mirror setting. The symptom here is installing process got stuck. So I track down the log file, and find out the real problem, the installation is looking for a file that's not there. This is just one of them, another pops up if I faked this file. This all happened during preseed, so I believe preseed has something to do with this. I google ubuntu preseed mirror and find this post saying: # If you select ftp, the mirror/country string does not need to be set. #d-i mirror/protocol string ftp d-i mirror/country string manual d-i mirror/http/hostname string archive.ubuntu.com d-i mirror/http/directory string /ubuntu d-i mirror/http/proxy string # Alternatively: by default, the installer uses CC.archive.ubuntu.com where # CC is the ISO-3166-2 code for the selected country. You can preseed this # so that it does so without asking. #d-i mirror/http/mirror select CC.archive.ubuntu.com # Suite to install. #d-i mirror/suite string lucid # Suite to use for loading installer components (optional). #d-i mirror/udeb/suite string lucid # Components to use for loading installer components (optional). #d-i mirror/udeb/components multiselect main, restricted I wonder the difference between d-i mirror/http/hostname and d-i mirror/http/mirror, I mean they all specify a mirror, right? In my preseed file, this is no d-i mirror/http/mirror, and d-i mirror/http/hostname points to my own repo as you might notice in the previous image. Here is my question: Does preseed fetches file/resource from internet, if I use local repo? Why it's looking for file that's not even there? This has bothered for quite time, many thanks in advance to anyone who might give any help.

    Read the article

  • Generic/Text Printer on Windows 7 not prompting for file name

    - by Trevor Tippins
    Hope someone can shed some light on this. I am downloading reports from an AIX-based system by directing them to a TT printer which the terminal emulator (MultiView 2000) intercepts and directs to the default printer on the local system. This local printer is configured as a vanilla Generic/Text printer attached to a FILE port. When I print from AIX, the output is spooled down and the local printer prompts for a file name into which to save the file...but not under Windows 7. This has worked fine for many years, on both Win2K and WinXP. However, on Windows 7 the output gets spooled as a file into spool\PRINTERS (and looks as expected) but the print job then hangs with a status of "Error - Printing" and never prompts for a file name. I have to cancel the job. The Generic/Text printer works as expected with other applications. I have tried setting the printer to print directly rather than spooling but this only serves to hang the terminal session too. I've also tried to run the emulator in Windows 2000 Compatibility Mode and as Administrator in case it was something like that but with no luck. As you might expect, it does work fine in XP Mode (as long as I print to a printer defined therein and not the host's printer) but operationally this isn't going to be an option. Obviously this emulation software is a decade old (at least) and I could just cross/upgrade all the users (at a cost) but, before I do so, has anyone seen this sort of behaviour before and found some sort of fix? Remote OS: AIX 5 Client OS: Windows 7 Pro (32-bit) Printer: Generic/Text on a FILE port TE Software: MultiView 2000 (32-bit) Thanks in advance.

    Read the article

  • Mercurial confusion - commit / push, backouts

    - by Madmanguruman
    I'm trying to set up a repository on a shared filesystem. I'm using Mercurial 2.1.2 on a Windows-based architecture. I start with an empty folder on the shared filesystem and create a repository in it. After this, I dump in the baseline files, and add them to versioning, then commit the changes. I then clone the repository to my local hard drive. I then make a change in my local repository, commit it, then push back to the shared filesystem repository. The shared repo graph I get in TortoiseHG looks strange (to me). This is the shared repo: This is the local repo: On the shared repo, the working directory always shows up on the top, then the graph goes 'down' to rev. 0 then back 'up' again through various revisions. It looks to me like I have two different branches, even though everything is on the default branch. Also, that 'top' revision always says "* Working Directory * Not a head revision!" I noticed that in my local repository, I don't get that dangling working directory at the top of the list - everything is in one branch. I also noticed that on my local repository, I can back out the tip revision with no problem. On the shared filesystem repository, I cannot, since I get an error ("Cannot backout change on a different branch"). How can this be? Aren't they supposed to be identical to each other? Am I fundamentally doing something wrong?

    Read the article

  • Boot records messed on dual boot (win7 and ubuntu) machine with SSD and HDD

    - by Michael
    i have a lenovo ideapad y570 with two hard drives: SSD and normal HDD both managed by RapidDrive and windows 7 pre-installed. First, i have shrunk my 500 GB HDD a little bit to make some place for a linux installation. Then i installed linux mint 12 to it, also installed grub onto the drive (dev/sdb). Installation programm has not allowed me to install grub on sda. Then i replaced linux mint with ubuntu 12.04 but installed grub onto the SSD (which is dev/sda and was the default-option). After that i could boot into my windows, only ubuntu worked. So i did a research, and tried: rewriting mbr of windows into sda1, reinstalling grub, replacing grub2 with grub-legacy, and now i think my partitions table are totally messed. Here is fdisk -l output: ubuntu@ubuntu:~$ sudo fdisk -l Disk /dev/sda: 64.0 GB, 64023257088 bytes 255 heads, 63 sectors/track, 7783 cylinders, total 125045424 sectors Units = sectors of 1 * 512 = 512 bytes Sector size (logical/physical): 512 bytes / 512 bytes I/O size (minimum/optimal): 512 bytes / 512 bytes Disk identifier: 0x00000000 Device Boot Start End Blocks Id System /dev/sda1 * 2048 411647 204800 7 HPFS/NTFS/exFAT /dev/sda2 411648 1009430959 504509656 7 HPFS/NTFS/exFAT Disk /dev/sdb: 500.1 GB, 500107862016 bytes 255 heads, 63 sectors/track, 60801 cylinders, total 976773168 sectors Units = sectors of 1 * 512 = 512 bytes Sector size (logical/physical): 512 bytes / 512 bytes I/O size (minimum/optimal): 512 bytes / 512 bytes Disk identifier: 0x5e5d1cc8 Device Boot Start End Blocks Id System /dev/sdb1 * 1979 884389887 442193954+ 12 Compaq diagnostics /dev/sdb2 884391934 976771071 46189569 5 Extended /dev/sdb5 884391936 937705471 26656768 83 Linux /dev/sdb6 937707520 967006207 14649344 83 Linux /dev/sdb7 967008256 976771071 4881408 82 Linux swap / Solaris I also cant mount any windows partitions to recover data. And when i open gparted, the whole sda-disk appears unallocated and it states "can not have a partition outside the disk!", also the end-sector address of /dev/sda2 confuses me. If i boot from the SSD, it throws some mbr error and wont boot, if i boot from the HDD, i only get the grub bash. How do i restore the partition tables? I can boot only from a live-cd at the machine. Thanks for any help.

    Read the article

  • How do I share a complete XP disk so it can be seen from a Windows 7 system? (To move all files to a

    - by Ian Ringrose
    This should be easier! (both computers can see the internet etc so I know the network it’s self is working) I have a normal home network with a Windows XP machine on it and the new Windows 7 (64 bit) machine. So I can transfer the files to the new Windows 7 machine, I wish to share the complete disk (and all files) from the Windows XP machine and access them from the Windows 7 machine. Is there a step by step set of instructions for doing this anywhere? So fare I have: put both computers into the same workgroup put the windows 7 machine into work network mode so it can see the XP machine in the work group shared the XP disk as read only But when I try to access a lot of the folders on the XP disks, I am told I am not allowed to access them. (I was not asked for any passwords by the windows 7 machine when I accessed the XP machine. The XP machine just has its default account with no password set on it) The XP machine runs XP home and hence has "simple file shairing" turn on. So it seems that even if I create a admin account (with password) and connect with that account, it still comes in as "guest" on the XP machine. Chooseing to share the folder I want access to rather then the top of the disk drive seems to work, but is a pain as I need to share each user's folder with a different share name. If the new computer was not a laptop, I would just plug the hard disk from the old machine into it, but being a laptop I don't have that option.

    Read the article

  • CentOS 6.5 new Kernel not active after reboot

    - by Kristofer
    Today I was running some yum updates and wanted to verify that everything went through fine by making sure I had a new kernel. To my surprise I noticed that CentOS was still running 2.6.32-431.5.1.el6.x86_64 even though it looked as though 2.6.32-431.23.3.el6 was installed. Indeed 2.6.32-431.23.3.el6 shows up in /etc/grub.conf but not in the upstart boot options. Any ideas why? In the update log it says: ---> Package kernel-firmware.noarch 0:2.6.32-431.5.1.el6 will be updated ---> Package kernel-firmware.noarch 0:2.6.32-431.23.3.el6 will be an update Could this be the reason? What does "will be an update" mean? My /etc/grub.conf: # grub.conf generated by anaconda # # Note that you do not have to rerun grub after making changes to this file # NOTICE: You have a /boot partition. This means that # all kernel and initrd paths are relative to /boot/, eg. # root (hd0,0) # kernel /vmlinuz-version ro root=/dev/mapper/VolGroup00-root # initrd /initrd-[generic-]version.img #boot=/dev/vda default=0 timeout=5 splashimage=(hd0,0)/grub/splash.xpm.gz hiddenmenu password --encrypted $1$auui(i$sODM4ni/Zts9IlMWu.wWF/ title CentOS (2.6.32-431.23.3.el6.x86_64) root (hd0,0) kernel /vmlinuz-2.6.32-431.23.3.el6.x86_64 ro root=/dev/mapper/VolGroup00-root rd_NO_LUKS LANG=en_US.UTF-8 KEYBOARDTYPE=pc KEYTABLE=sv-latin1 rd_NO_MD rd_LVM_LV=VolGroup00/swap SYSFONT=latarcyrheb-sun16 crashkernel=auto rd_LVM_LV=VolGroup00/root rd_NO_DM rhgb quiet rhgb quiet audit=1 initrd /initramfs-2.6.32-431.23.3.el6.x86_64.img title CentOS (2.6.32-431.5.1.el6.x86_64) root (hd0,0) kernel /vmlinuz-2.6.32-431.5.1.el6.x86_64 ro root=/dev/mapper/VolGroup00-root rd_NO_LUKS LANG=en_US.UTF-8 KEYBOARDTYPE=pc KEYTABLE=sv-latin1 rd_NO_MD rd_LVM_LV=VolGroup00/swap SYSFONT=latarcyrheb-sun16 crashkernel=auto rd_LVM_LV=VolGroup00/root rd_NO_DM rhgb quiet rhgb quiet audit=1 initrd /initramfs-2.6.32-431.5.1.el6.x86_64.img title CentOS (2.6.32-431.el6.x86_64) root (hd0,0) kernel /vmlinuz-2.6.32-431.el6.x86_64 ro root=/dev/mapper/VolGroup00-root rd_NO_LUKS LANG=en_US.UTF-8 KEYBOARDTYPE=pc KEYTABLE=sv-latin1 rd_NO_MD rd_LVM_LV=VolGroup00/swap SYSFONT=latarcyrheb-sun16 crashkernel=auto rd_LVM_LV=VolGroup00/root rd_NO_DM rhgb quiet rhgb quiet audit=1 initrd /initramfs-2.6.32-431.el6.x86_64.img

    Read the article

  • Windows 7 Blank Screen on Boot / Login

    - by Greg
    I have a new system that's having a few problems... sometimes (seems to be when the PC is cold, i.e. has been switched off for a while, though that could be my imagination) I get a blank blue screen when I boot up. The system boots normally and auto-logs-in. The desktop loads and I'm even able to launch applications, but then everything disappears and the screen goes to the default windows desktop blue colour (not the desktop image, just a plain blue with no mouse cursor). At this point the machine completely locks up - I'm unable to even toggle Num Lock and have to hold in the power button for 5 seconds to kill it. Interestingly if I manage to launch some applications before it goes blank, they will usually crash... sometimes explorer.exe will crash too. When I reboot, the system is fine and stable. I've installed the latest graphics drivers and run memtest86+ for 6 passes (and counting) with no errors. The system specs are: CPU: Intel I7 2.66 @ 3.4GHz RAM: 6GB (3 * 2GB DDR3) HDD: 128GB Crucial M225 SSD Motherboard: Gigabyte EX58-UD3R Gfx: ATI Radeon Sapphire 5870 1GB Note: There are a few similar questions but I haven't found one that matches my symptoms

    Read the article

< Previous Page | 869 870 871 872 873 874 875 876 877 878 879 880  | Next Page >