Search Results

Search found 7154 results on 287 pages for 'networking'.

Page 139/287 | < Previous Page | 135 136 137 138 139 140 141 142 143 144 145 146  | Next Page >

• Google account gives ERR_SSL_BAD_RECORD_MAC_ALERT errors

  - by Kjensen

  A couple of days ago, I started being unable to connect to accounts.google.com, which handles logins to all kinds of google services. I get this error in Chrome: Error 126 (net::ERR_SSL_BAD_RECORD_MAC_ALERT): Unknown error. In IE I get this: I assume it is the same error, just wrapped up. I run Win8 RTM. On the SAME machine, using the same network card, in a VMWare workstation image running Win7, I am able to connect perfectly. On another of my machines on my network, I am also still able to connect with no problem. My girlfriend uses the same network and has also complained a couple of times about this error (google calendar) - but this is anecdotal, since her technical troubleshooting abilities stop at "xxxx is broken". Her machine runs Win7. ;) I have rebooted, cleared cookies, do not run any antivirus/firewall, have not changed network config. The first 3-4 days after installing Win8, I did not have any problems. I have also searched, and found a hint about enabling SSL2.0 in connection settings, which did not help. Anybody know something about this error and what I can do to fix it?

  Read the article

• Tunneling a public IP to a remote machine

  - by Jim Paris

  I have a Linux server A with a block of 5 public IP addresses, 8.8.8.122/29. Currently, 8.8.8.122 is assigned to eth0, and 8.8.8.123 is assigned to eth0:1. I have another Linux machine B in a remote location, behind NAT. I would like to set up an tunnel between the two so that B can use the IP address 8.8.8.123 as its primary IP address. OpenVPN is probably the answer, but I can't quite figure out how to set things up (topology subnet or topology p2p might be appropriate. Or should I be using Ethernet bridging?). Security and encryption is not a big concern at this point, so GRE would be fine too -- machine B will be coming from a known IP address and can be authenticated based on that. How can I do this? Can anyone suggest an OpenVPN config, or some other approach, that could work in this situation? Ideally, it would also be able to handle multiple clients (e.g. share all four of spare IPs with other machines), without letting those clients use IPs to which they are not entitled.

  Read the article

• SSH not working after Restoring Running-Config to a Replacement Cisco Router

  - by Kyle Brandt

  One of my Cisco routers died over the weekend, Cisco sent the replacement and I restored the the config using copy tftp: running-config. Everything seems to work fine but I can no longer ssh into the router (I can telnet). The connection is refused, so it isn't listening on port 22 it seems like. I had previously backed up the config by just doing ssh router 'show run' > backup_config from my workstation. So: Is there anything wrong with my method of backup vs copy running-config tftp:? I know I haven't given any debug information, but is there something typical I need to do to get ssh working?

  Read the article

• Doing port forwarding and then using it from within the internal network

  - by Ram Rachum

  We all know that by doing port forwarding on the router, computers from outside the network are able, on the specified ports, to access internal computers by targeting the external IP. I'm now replacing a TP-Link router with a D-link VDSL N 6740U router, (and copied over all the settings,) and I've noticed that one thing stopped working: With the TP-link router, you could access those port-forwarded computers from within the network, using the external IP, and they would be forwarded to the relevant computers. With the new D-Link router, it doesn't work. You might be wondering, why would you want to use the external IP and port forwarding when you're inside the internal network anyway and can just access the internal IP? One example for why this is useful: You have an iPhone app that connects to a service on an internal computer. The iPhone app knows to connect to the external IP. When we put that iPhone inside the internal network (via WiFi), it suddenly stops working, because it can't access the service from the external IP anymore. Is it an inherent property of D-Link routers that they do not allow accessing internal servers from inside the network by targeting the external IP? Or is there a way to make it work?

  Read the article

• "Play on another device" function not working with Samsung Allshare

  - by Tural Teyyuboglu

  Briefly I can't get work "Play on another device" function More detailed My Samsung TV (which support network) and PC are in the same network. The problem is, I can play PC contents from TV. But when I try to control TV remotely from Allshare software (PC), it shows TV only as connected device: But not as remote player in "players" list The result Tried to turn off firewall completely, reset router, re-install software. No success. Please help. BTW. On this link they kinda explained how to do it, but I software doesn' detect the TV as player in my case: http://www.samsung.com/global/allshare/pcsw/quickguide.html

  Read the article

• Trixbox CentOS Default GW Problem (Multi-homed server)

  - by slashp

  I'm having an issue with a CentOS trixbox server which is dual-homed (one private facing NIC [eth1], one internet-facing NIC [eth0]). I can't seem to get the default gateway to set properly to our ISP's GW via eth0. I've modified the /etc/sysconfig/network to contain both a GATEWAY & GATEWAYDEV line and removed the GATEWAY line from /etc/sysconfig/network-scripts/ifcfg-eth1 (as well as /etc/sysconfig/network-scripts/ifcfg-eth0). No default GW shows up in the routing table unless it's specified in the ifcfg-eth1 file (which both the wrong interface and wrong gateway IP), otherwise, the routing table simply does not contain a default gateway..any ideas would be greatly appreciated! Thanks! EDIT Just realized when attempting to add the default gateway manually using the route add command, I receive an error stating: SIOCADDRT: Network is unreachable I know this error can occur when your default gateway and interface IP address are not on the same subnet..in this case, my public IP address of eth0 is a /29.

  Read the article

• Tomato/DD-WRT router to act as switch & only NAT some port

  - by fseto

  BACKGROUND: I have a device that must use a real IP address. Currently, my ISP uses DHCP and I can have up to 4 real IP address assigned. However, the cable modem only have 1 ethernet port and it's connected to my router (running Tomato, but can run DD-wrt or other Openwrt if required). Question stems from how I can connect the additional device, requiring a real IP? EASY SOLUTION: would be to get a switch and connect to the CM, Router, and Device. But alas, I want to avoid this route, since: my wiring cabinet in my home is drawing lots of power and heat already Device will be unprotected by any firewall unable to monitor the traffic to/from device. Besides, what would be the FUN in that? =) IDEA: So what I want to do is to configure the router, so that one of the switchport is removed from the normal br0 bridge. Instead, I want to make it behave like a switch on the WAN port. What's the best way of doing this? Should I create another bridge on the WAN & the device port? Can a single port belongs to two bridges? or would I need to create a subinterface first? Would I need a DHCP-relay? Am I expecting too much from my poor cheapie router? +------+ | CM | +--++--+ || +----WAN---------------+ | / \ Router | | BR1? BR0 | | | \ | | | {NAT} | | | / | | \ | +-P0----P1-P2-P3-Wifi--+ | +------+ |Device| +------+

  Read the article

• Dual-WAN router

  - by aix

  I am looking for a router that would fit the following requirements: Two WAN interfaces: the primary is PPPoE, the secondary will link to a GigE port on another router (a 100Mbps link will suffice); Two (ideally four) GigE LAN ports; No requirement for a firewall; No requirement for Wi-Fi; Inexpensive. The plan for the two WAN interfaces is as follows. All outbound traffic will go to the primary, with exceptions based on destination IP/subnet or possibly on src+dest IPs/subnets. Such exceptions should be routed to the secondary. It would be very nice if, should the primary go down, the secondary would automatically take over for all outbound traffic. I am reasonably sure that I can put something together based on dd-wrt. However, I'd like to hear from you what alternatives are out there (especially something easier to set up for my use case, even if it means paying more for the hardware.)

  Read the article

• How to force Windows 7 to ask for a "fresh" IP address from DHCP server?

  - by haimg

  I'm troubleshooting a certain issue with my DHCP configuration, and need my Windows machine to ask for a "fresh" IP address, so I can see which address DHCP server gives by default. When I do ipconfig /release followed by ipconfig /renew, Windows "proposes" its old IP address to the DHCP server (just checked with Wireshark, initial "DHCP Discover" message has Option-50 (requested IP address) with Windows machine's old IP). Tried disabling/enabling network adapter. Same behavior. Question: How can I force Windows to just ask for a new IP address, without proposing its old IP address.

  Read the article

• How can I switch from a custom linux network namespace back to the default one?

  - by Martin

  With ip netns exec you can execute a command in a custom network namespace - but is there also a way to execute a command in the default namespace? For example, after executing these two commands: sudo ip netns add test_ns sudo ip netns exec test_ns bash How can the newly created bash execute programs in the default network namespace? There is no ip netns exec default or anything similar as far as I've found. My scenario is: I want to run a SSH server in a separate network namespace (to keep the rest of the system unaware of the network connection, as the system is used for network testing), but want to be able to execute programs in the default network namespace via the SSH connection. What I've found out so far: Created network namespaces are listed as files under /var/run/netns (but there is no file for the default namespace) The ip netns exec code can be found here: http://git.kernel.org/cgit/linux/kernel/git/shemminger/iproute2.git/tree/ip/ipnetns.c#n132 - I haven't grasped everything that it is doing yet, but it doesn't look very promising. ip netns identify $$ as suggested by Howto query and change network namespace on linux? returns nothing when in the default network namespace

  Read the article

• How to determine most stable Cisco IOS release?

  - by Chris J

  This post is about a Catalyst 4948E switch. I was looking on the download page and realized that there are no "GD" versions available. Are the "ED" versions stable? Even if you change "ED" to "GD" in the URL the IOS images are still the same. http://www.cisco.com/cisco/software/release.html?mdfid=283027810&flowid=3592&softwareid=280805680&release=15.1.1-SG2&relind=AVAILABLE&rellifecycle=ED&reltype=latest Is 15.1 as reliable as 15.0? My devices are currently on the 12.2 train. Is there anything special to upgrade to one of the 15.x trains? Are the configurations compatible.

  Read the article

• PowerConnect 3548p SNTP and web interface not working

  - by Force Flow

  I have been unable to get SNTP and access to the web interface working properly on a Dell PowerConnect 3548p. In the logs, this message appears over and over again: 04-Jan-2000 20:19:29 :%MNGINF-W-ACL: Management ACL drop packet received on interface Vlan 172 from 172.17.0.3 to 172.18.0.10 protocol 17 service Snmp 172 is the management vlan. 172.17.0.3 is the DNS server 172.18.0.10 is the switch's IP address. The DNS server and the switch are located on different subnets and separated by routers. I am unable to access the web interface of the switch from the 172.17.x.x subnet. I can only access the web interface of the switch if I am accessing it from the 172.18.x.x subnet. There is also a managed linksys switch on the 172.18.x.x subnet on the 172 vlan, which has no problem with SNTP. I can also access it from the 172.17.x.x network. So, it stands to reason that this is not a firewall or routing issue, but with the 3548p switch. I suspect the issue is with management permissions/ACLs on the 3348p switch, but that's about as much as I've been able to determine so far. Any ideas?

  Read the article

• Assign fixed IP address via DHCP by DNS lookup

  - by Janoszen

  Preface I'm building a virtualization environment with Ubuntu 14.04 and LXC. I don't want to write my own template since the upgrade from 12.04 to 14.04 has shown that backwards compatibility is not guaranteed. Therefore I'm deploying my virtual machines via lxc-create, using the default Ubuntu template. The DNS for the servers is provided by Amazon Route 53, so no local DNS server is needed. I also use Puppet to configure my servers, so I want to keep the manual effort on the deployment minimal. Now, the default Ubuntu template assigns IP addresses via DHCP. Therefore, I need a local DHCP server to assign IP addresses to the nodes, so I can SSH into them and get Puppet running. Since Puppet requires a proper DNS setup, assigning temporary IP addresses is not an option, the client needs to get the right hostname and IP address from the start. Question What DHCP server do I use and how do I get it to assign the IP address based only on the host-name DHCP option by performing a DNS lookup on that very host name? What I've tried I tried to make it work using the ISC DHCP server, however, the manual clearly states: Please be aware that only the dhcp-client-identifier option and the hardware address can be used to match a host declaration, or the host-identifier option parameter for DHCPv6 servers. For example, it is not possible to match a host declaration to a host-name option. This is because the host-name option cannot be guaranteed to be unique for any given client, whereas both the hardware address and dhcp-client-identifier option are at least theoretically guaranteed to be unique to a given client. I also tried to create a class that matches the hostname like this: class "my-client-name" { match if option host-name = "my-client-name"; fixed-address my-client-name.my-domain.com; } Unfortunately the fixed-address option is not allowed in class statements. I can replace it with a 1-size pool, which works as expected: subnet 10.103.0.0 netmask 255.255.0.0 { option routers 10.103.1.1; class "my-client-name" { match if option host-name = "my-client-name"; } pool { allow members of "my-client-name"; range 10.103.1.2 10.103.1.2; } } However, this would require me to administer the IP addresses in two places (Amazon Route53 and the DHCP server), which I would prefer not to do. About security Since this is only used in the bootstrapping phase on an internal network and is then replaced by a static network configuration by Puppet, this shouldn't be an issue from a security standpoint. I am, however, aware that the virtual machine bootstraps with "ubuntu:ubuntu" credentials, which I intend to fix once this is running.

  Read the article

• Why does my Intel Tolapai network chip not transmit packets?

  - by Hanno Fietz

  I'm trying to deploy an embedded system (NISE 110 by Nexcom) based on the Intel EP80579 (Tolapai) chip. Tolapai apparently integrates controllers for Ethernet etc. on a single chip (Intel homepage). The machine can't get a network connection. Diagnosis as far as I could manage: Drivers drivers from Intel compiled and installed without problems (version 1.0.3-144). Kernel version and Linux distribution (CentOS 5.2, 2.6.18) match the driver's installation instructions. drivers are loaded and show up in lsmod (module names are gcu and iegbe) interfaces eth0 and eth1 show up in ifconfig ifconfig I can bring up the interfaces with fixed IP pinging the interface locally works ifconfig shows flag UP but not RUNNING Link ethtool shows "Link detected: no", "Speed: unknown (65536)" and "Duplex: unknown (255)" Link LED is on on the other side of the cable, ethtool shows "Link detected: yes" and reports a speed of 1000 Mbps, which has allegedly been auto-neogotiated with the problematic device. Network traffic analysis the device does not reply on ARP, ICMP echo or anything else (iptables is down) when trying to send ICMP or DHCP requests, they never reach the other end activity LED is off on the device, on at the other end. I tried the following without any effect: Different cables (2 straight, one crossed), I get the link LED lit up on each. Three different devices on the other end (one PC, one netbook, one router) Fixed ARP table entries on both sides Connecting both network ports of the machine with each other, won't ping through the cable, but will ping locally. Tried straight and crossed cables for that.

  Read the article

• Under FreeBSD, can a VLAN interface have a smaller MTU than the primary interface?

  - by larsks

  I have a system with two physical interfaces, combined into a LACP aggregation group. That LACP channel has two VLANs, one untagged (the "native vlan") and one using VLAN tagging. This gives us: lagg0: flags=8843<UP,BROADCAST,RUNNING,SIMPLEX,MULTICAST> metric 0 mtu 1500 options=19b<RXCSUM,TXCSUM,VLAN_MTU,VLAN_HWTAGGING,VLAN_HWCSUM,TSO4> ether 00:25:90:1d:fe:8e inet 10.243.24.23 netmask 0xffffff00 broadcast 10.243.24.255 media: Ethernet autoselect status: active laggproto lacp laggport: em1 flags=1c<ACTIVE,COLLECTING,DISTRIBUTING> laggport: em0 flags=1c<ACTIVE,COLLECTING,DISTRIBUTING> vlan0: flags=8843<UP,BROADCAST,RUNNING,SIMPLEX,MULTICAST> metric 0 mtu 1500 options=3<RXCSUM,TXCSUM> ether 00:25:90:1d:fe:8e inet 10.243.16.23 netmask 0xffffff80 broadcast 10.243.16.127 media: Ethernet autoselect status: active vlan: 610 parent interface: lagg0 Is it possible to set a 9K MTU on lagg0 while preserving the 1500 byte MTU on vlan0? Normally I would simply try this out, but this is actually on a vendor-supported platform and I am loathe to make changes "behind the back" of their administration interface. This system is roughly FreeBSD 7.3.

  Read the article

• My DNS works! But, what is the simplest way to add something to it?

  - by Alex

  This is my current DNS example.com.db zone file. I followed a tutorial. It works, because when I point to this DNS from another server via resolve.conf, it will actually forward me to the right IP when I do "ping example.com". ; ; BIND data file for example.com ; $TTL 604800 @ IN SOA example.com. info.example.com. ( 2007011501 ; Serial 7200 ; Refresh 120 ; Retry 2419200 ; Expire 604800) ; Default TTL ; @ IN NS ns1.example.com. @ IN NS ns2.example.com. example.com. IN MX 10 mail.example.com. example.com. IN A 192.168.254.1 www IN CNAME example.com. mail IN A 192.168.254.1 ftp IN CNAME example.com. example.com. IN TXT "v=spf1 ip4:192.168.254.1 a mx ~all" mail IN TXT "v=spf1 a -all" Right now, ping example.com....goes to 192.168.254.1. That's great!!! it works! My question is--how can I add something do this file so that when my other servers: ping dbserver1....goes to 44.245.66.222 ping cacheserver1 ....goes to 38.221.44.555 I want to use it like a universal hosts file for my machines.

  Read the article

• Torrent upload ratio not updated on Synology DS212+

  - by user179271

  I have a Synology DS212+ NAS running DSM 4.2-3211 (current version). I use it for several purposes including torrent download using Download Station and a tracker that needs authentication. My problem is that my download/upload ratio isn't updated, so it constantly falls down. My NAS is behind a router, and I configured the NAT to forward ports 6890 to 6999 to the internal IP address of the NAS. Here are the Download Station settings : TCP port : 6990, Sharing ratio : 900%, Sharing time : infinite, max download speed : 0 (no limit), max upload speed : 0 (no limit), BT protocol encryption : checked, max numbers of peers allowed by torrent file : 4000, DHT : checked, with port 6889. When the DHT option is not checked, the NAS doesn't upload any files. I don't know what is this option for. Can someone help me to solve this problem ? Did I miss any step, or does it come from the NAT ? How is the authentication managed by Dowload Station ? (Sorry for my english) Thanks.

  Read the article

• Understanding tcptraceroute versus http response

  - by kojiro

  I'm debugging a web server that has a very high wait time before responding. The server itself is quite fast and has no load, so I strongly suspect a network problem. Basically, I make a web request: wget -O/dev/null http://hostname/ --2013-10-18 11:03:08-- http://hostname/ Resolving hostname... 10.9.211.129 Connecting to hostname|10.9.211.129|:80... connected. HTTP request sent, awaiting response... 200 OK Length: unspecified [text/html] Saving to: ‘/dev/null’ 2013-10-18 11:04:11 (88.0 KB/s) - ‘/dev/null’ saved [13641] So you see it took about a minute to give me the page, but it does give it to me with a 200 response. So I try a tcptraceroute to see what's up: $ sudo tcptraceroute hostname 80 Password: Selected device en2, address 192.168.113.74, port 54699 for outgoing packets Tracing the path to hostname (10.9.211.129) on TCP port 80 (http), 30 hops max 1 192.168.113.1 0.842 ms 2.216 ms 2.130 ms 2 10.141.12.77 0.707 ms 0.767 ms 0.738 ms 3 10.141.12.33 1.227 ms 1.012 ms 1.120 ms 4 10.141.3.107 0.372 ms 0.305 ms 0.368 ms 5 12.112.4.41 6.688 ms 6.514 ms 6.467 ms 6 cr84.phlpa.ip.att.net (12.122.107.214) 19.892 ms 18.814 ms 15.804 ms 7 cr2.phlpa.ip.att.net (12.122.107.117) 17.554 ms 15.693 ms 16.122 ms 8 cr1.wswdc.ip.att.net (12.122.4.54) 15.838 ms 15.353 ms 15.511 ms 9 cr83.wswdc.ip.att.net (12.123.10.110) 17.451 ms 15.183 ms 16.198 ms 10 12.84.5.93 9.982 ms 9.817 ms 9.784 ms 11 12.84.5.94 14.587 ms 14.301 ms 14.238 ms 12 10.141.3.209 13.870 ms 13.845 ms 13.696 ms 13 * * * … 30 * * * I tried it again with 100 hops, just to be sure – the packets never get there. So how is it that the server does respond to requests via http, even after a minute? Shouldn't all requests just die? I'm not sure how to proceed debugging why this server is slow (as opposed to why it responds at all).

  Read the article

• SCP command Clarification

  - by david.colais

  I'm using the scp commands to pull some files from the remote server and one variation of the command is not working. I have 2 files names one.xml and two.xml in a remote server and I'm pulling these two files into the current dir using the following command: scp [email protected]:/student/class/Intermediate/one.xml . scp [email protected]:/student/class/Intermediate/two.xml . The above command works fine but if I use wildcards to pull all the xml files in a single shot as shown below it returns scp: No match. scp [email protected]:/student/class/Intermediate/*.xml . Why is it working if I pull the files individually and not working if I try to pull using wildcards.

  Read the article

• Monitor mode 802.11 captures on OSX

  - by Mike A

  I'm trying to determine the difference between capturing 802.11 frames in the following ways on OSX (10.8.5). It's a bit esoteric, but I use "Option 2" to capture frames for later analysis, and am wondering if I'm missing something. Option 1: use "airportd": $sudo /usr/libexec/airportd en0 sniff Option 2: use "airport" followed by tcpdump: sudo /System/Library/PrivateFrameworks/Apple80211.framework/Versions/Current/Resources/airport --channel= sudo tcpdump -I -P -i en0 -w /tmp/capture.pcap (or alternatvely eliminate the -w and watch packets real-time). From what I can tell: Both commands, according to the wifi icon on OSX, put the interface into 'monitor' mode. Both commands output a pcap file that is readable in both wireshark/tcpdump & Eye PA. Both commands appear to capture management, control and data frames. The rub: Option 1 disconnects you from the network. This is expected, when putting an interface into 'monitor' mode. Option 2 does NOT disconnect you, provided you've set the channel to the same channel your currently connected to. This has a distinct advantage of keeping your connection up while capturing in monitor mode. My question: Option 2 does not seem like it should work, or more specifically, it does not seem like I should be able to remain connected while also capturing frames in monitor mode. On a wired NIC, you can be 'promiscuous' and still send frames, though I didn't think the same was true for wireless NIC. I'm questioning the validity of capturing frames w/ Option 2?

  Read the article

• Linux/hostapd: AP can ping clients, clients can access internet, can't access www@wlan1 with more than 5-6 packets at once

  - by mhambra

  Please edit the title, can't make it sound better. -- OP. Hi all, I have a Wifi USB dongle in a PC, that serves as an AP for laptop. wlan1: 192.168.2.1, netmask 255.255.255.0, routed: route add -net 192.168.2.0 netmask 255.255.255.0 gw 192.168.2.1 ping 192.168.2.2 (laptop): ping was ok for lot of packets. Now, I try to access 192.168.2.1:80/myindex.html (apache) from laptop, and can see that own 1kb test page. But, trying to access 192.168.2.1:80/my.jpg, I see the following: GET /my.jpg HTTP/1.1 200 OK <jpg header, about a kilobyte> <TCP packet retransmisson> <TCP packet retransmisson> <end of stream> It seems to be a hostapd's problem (networked stuff worked fine with Ad-Hoc), but it may be also forwarding/routing problem too. What to google for? Even more strange, SSH to that host works fine.

  Read the article

• why udp client work when wirshark capture?

  - by herzl shemuelian

  I have two machine A,B windows 7 os .I connect them end to end and try run a performance test by using tcpreplay. step 1) I check conectivity between to point by netcat In A i run nc -lvup 5432 when I run on B nc -u 1.2.3.4 5432 I can send data from B to A step 2) when in I run tcpreplay in B tcpreplay -i %0 myudp.pcap in A I don't recevice any data . when I open wireshark in A then my nc can read data why? I check dst mac and dst ip in pcap file they are correct. is importan udp src mac or src ip for udp how that I open udp server ?

  Read the article

• Win 2003 STD network adapter always showing DHCP when in static IP configuration, + it loses the DNS

  - by Darragh

  Hi, I have a server that after the first configuration it was DHCP, now I have added it to our domain and in a static IP, however after a few moments it returns to DHCP but with only some of the IPv4 setting staying the same, It loses DNS for example. I'm not sure what is causing the problem but all I know is this started to happen after I added it to the domain, Would it be a domain policy? or the NIC drivers Spec; Dell M605 Blade server Windows 2003 STD SP1 Intel Xeon Quad core NIC: Dual embedded Broadcom NetXtreme IITM 5708 Gigabit Ethernet NIC w/ TOE

  Read the article

• Transient network dropout for Xen DomU's

  - by Stephen C

  We've got a CentOS server running a cluster of virtuals. Occasionally the cluster's internal network drops out for a minute or so ... and then comes back. The problem is somehow related to the actual network traffic, but it is not a simple load issue. (The system is generally lightly loaded, and the problem occurs irrespective of actual load.) The setup: CentOS 5.6 on Dom0, various CentOS on the DomU's Hardware - a Dell R710 with a BroadCom NextXpress 2 NIC (sigh) using the latest drivers for the NIC from BroadCom Xen configured to use network-bridge and vif-bridge Some iptable tweaks to route an unrelated port to one of the virtuals. The system has one externally visible IP address, and Dom0 runs an Apache httpd configured with a number of virtual hosts each of which reverse proxies to web servers running on the virtuals. (The virtuals have to be NAT'ed, primarily because we don't have enough allocated public IP addresses.) The symptoms: Works fine most of the time. When someone tries to UPLOAD a large file to one virtuals, the internal network drops out ... for all virtuals: The Dom0 httpd sees a network timeout talking to the backend server on the virtual and reports a 502. A previously established ssh connection from Dom0 to any of the DomU's freezes. Our monitoring shows ping failures for traffic between virtuals. The Xen consoles to the DomU's do not freeze. No log messages in any log files that I can see, on either Dom0 or the DomU's ... apart from the Dom0 httpd logs. After a minute or so, the problem clears by itself. This is 100% reproducible. What we've tried: Downloading, building and installing the latest BNX2 driver on Dom0 Turning off MSI on the NIC - adding "options bnx2 disable_msi=1" to /etc/modprobe.conf Turning off tcp segmentation offload - "ethtool -K eth0 tso off". Sacrificing a black rooster at midnight. I've exhausted all my options apart from switching to KVM ... or slaughtering more roosters. Any suggestions?

  Read the article

• How do I keep a bridge enabled on a bonded interface?

  - by jlawer

  I'm working on setting up a pair of CentOS 6.3 servers that will run a couple of KVM vms and have come across a problem setting up a bridge on a bond. I am using Mode 4 (802.3ad) bonding on a pair of stacked Dell Powerconnect 5524 switches connecting to R320 servers. There are 2 links (1 to each switch) that form a Link Aggregation Group (802.3ad / LACP bonding). On top of the bond I have VLAN Tagging. I've verified this is a problem on multiple other bonding modes so it isn't just a mode 4 issue. I am testing what happens when 1 link is dropped (ie switch dies, cable breaks, etc). If I don't have a bridge (for KVM), everything works fine, failover happens as expected. If I have the bridge enabled, it works fine until failover (unplugging a cable). When failover happens /var/log/messages shows the slave link going down, followed within a second by: kernel: br1: port 1(bond0.8) entering disabled state The thing is /proc/net/bonding/bond0 shows the link is up as expected (simply with only 1 slave instead of 2). If I plug the cable back in it recovers and brings the bridge back to an enabled state. I actually have tested this while a ping is occuring and if the timing is right a packet will actually leave the system after the link is lost, but before the disabled message occurs. This disabled state I assumed was STP, but I have disabled STP on the bridge configuration and this issue still occurs. brctl showstp br1 still shows the link as disabled when it is running without a slave. I also switched between the nics in the server (I have 2x Broadcom & 4x intel). It doesn't matter which configuration I have. Does anyone know of a way to force the bridge to stay enabled or why its detecting the bond as disabled, when it isn't?

  Read the article

< Previous Page | 135 136 137 138 139 140 141 142 143 144 145 146  | Next Page >