Search Results

Search found 4274 results on 171 pages for 'references'.

Page 146/171 | < Previous Page | 142 143 144 145 146 147 148 149 150 151 152 153  | Next Page >

  • Trying to overload + operator

    - by FrostyStraw
    I cannot for the life of me understand why this is not working. I am so confused. I have a class Person which has a data member age, and I just want to add two people so that it adds the ages. I don't know why this is so hard, but I'm looking for examples and I feel like everyone does something different, and for some reason NONE of them work. Sometimes the examples I see have two parameters, sometimes they only have one, sometimes the parameters are references to the object, sometimes they're not, sometimes they return an int, sometimes they return a Person object. Like..what is the most normal way to do it? class Person { public: int age; //std::string haircolor = "brown"; //std::string ID = "23432598"; Person(): age(19) {} Person operator+(Person&) { } }; Person operator+(Person &obj1, Person &obj2){ Person sum = obj1; sum += obj2; return sum; } I really feel like overloading a + operator should seriously be the easiest thing in the world except I DON'T KNOW WHAT I AM DOING. I don't know if I'm supposed to create the overload function inside the class, outside, if it makes a difference, why if I do it inside it only allows one parameter, I just honestly don't get it.

    Read the article

  • Real time content editing html5

    - by Mark Lauzon
    So I've seen things like WordPress and FCKEditor, and basically a bunch of stuff that uses external code that I can't see or edit. Whenever I ask about editing and saving the content of a page in real time I just get referenced to an API or I get handed code that only changes the page until it's reloaded. What I want to know is how do I code it myself? I want to add real time content editing to a page without the use of someone else's code. I've checked out code for various forums and wikipedia and whatnot, and all of it references code I don't have access to. Is this a thing? Can I edit a page in real time? I thought of writing the edited text to a file on the server, and then when they click save, reading it back into the code to the section they were editing, but I don't know how to do that or if it's even possible. As a side note, I'm very new to html, but not new to coding. EDIT: The structure can be very much like Wikipedia, it doesn't have to be real time, it just has to work

    Read the article

  • int foo(type& bar); is a bad practice?

    - by Earlz
    Well, here we are. Yet another proposed practice that my C++ book has an opinion on. It says "a returning-value(non-void) function should not take reference types as a parameter." So basically if you were to implement a function like this: int read_file(int& into){ ... } and used the integer return value as some sort of error indicator (ignoring the fact that we have exceptions) then that function would be poorly written and it should actually be like void read_file(int& into, int& error){ } Now to me, the first one is much clearer and nice to use. If you want to ignore the error value, you do so with ease. But this book suggests the later. Note that this book does not say returning value functions are bad. It rather says that you should either only return a value or you should only use references. What are your thoughts on this? Is my book full of crap? (again)

    Read the article

  • How to start networking on a wired interface before logon in Ubuntu Desktop Edition

    - by Burly
    Problem Ubuntu 9.10 Desktop Edition (and possibly previous versions as well, I haven't tested them) has no network connections after boot until at least 1 user logs in. This means any services that require networking (e.g. openssh-server) are not available until someone logs in locally either via gdm, kdm, or a TTY. Background Ubuntu 9.10 Desktop Edition uses the NetworkManager service to take commands from the nm-applet in Gnome (or it's equivalent in KDE). As I understand it, while NetworkManager is running at boot, it is not issued any commands to connect until you login for the first time because nm-applet isn't running until you login and your Gnome session starts (or similar for KDE). I'm not sure what prompts NetworkManager to connect to the network when you login via a TTY. There are several relevant variables involved in starting up the network connections including: Wired vs Wireless (and the resulting drivers, SSID, passwords, and priorities) Static vs DHCP Multiple interfaces Constraints Support Ubuntu 9.10 Karmic Koala (bonus points for additional supported versions) Support wired eth0 interface Receive an IP address via DHCP Receive DNS information via DHCP (obviously the DHCP server must provide this information) Enable networking at the proper time (e.g. some time after file systems are loaded but before network services like ssh start) Switching distros or versions (e.g. to Server Edition) is not an acceptable solution Switching to a Static IP configuration is not an acceptable solution Question How to start networking on a wired interface before logon in Ubuntu Desktop Edition? What I have tried Per this guide, adding the following entry into /etc/network/interfaces so that NetworkManager won't manage the eth0 interface: auth eth0 iface inet dhcp After reboot eth0 is down. Issuing ifconfig eth0 up brings the interface up but it receives no IP address. Issuing dhclient eth0 instead Does bring up the interface and it Does receive an IP address. Completely removing the NetworkManager package in addition to the settings above. I'm a bit confused with the whole UpStart/SysVinit mangling that's going in Ubuntu currently (I'm more familiar with the CentOS world). However, directly issuing sudo /etc/init.d/networking start Or sudo start networking does not bring up the eth0 interface at all, much less get an IP address. See-Also How to force NetworkManager to make a connection before login? References Ubuntu Desktop Edition Ubuntu Networking Configuration Using Command Line Automatic Network Configuration Via Command-Line Start network connection before login

    Read the article

  • iptables syn flood countermeasure

    - by Penegal
    I'm trying to adjust my iptables firewall to increase the security of my server, and I found something a bit problematic here : I have to set INPUT policy to ACCEPT and, in addition, to have a rule saying iptables -I INPUT -i eth0 -j ACCEPT. Here comes my script (launched manually for tests) : #!/bin/sh IPT=/sbin/iptables echo "Clearing firewall rules" $IPT -F $IPT -Z $IPT -t nat -F $IPT -t nat -Z $IPT -t mangle -F $IPT -t mangle -Z $IPT -X echo "Defining logging policy for dropped packets" $IPT -N LOGDROP $IPT -A LOGDROP -j LOG -m limit --limit 5/min --log-level debug --log-prefix "iptables rejected: " $IPT -A LOGDROP -j DROP echo "Setting firewall policy" $IPT -P INPUT DROP # Deny all incoming connections $IPT -P OUTPUT ACCEPT # Allow all outgoing connections $IPT -P FORWARD DROP # Deny all forwaring echo "Allowing connections from/to lo and incoming connections from eth0" $IPT -I INPUT -i lo -j ACCEPT $IPT -I OUTPUT -o lo -j ACCEPT #$IPT -I INPUT -i eth0 -j ACCEPT echo "Setting SYN flood countermeasures" $IPT -A INPUT -p tcp -i eth0 --syn -m limit --limit 100/second --limit-burst 200 -j LOGDROP echo "Allowing outgoing traffic corresponding to already initiated connections" $IPT -A OUTPUT -p ALL -m state --state ESTABLISHED,RELATED -j ACCEPT echo "Allowing incoming SSH" $IPT -A INPUT -p tcp --dport 22 -m state --state NEW -m recent --set --name SSH -j ACCEPT echo "Setting SSH bruteforce attacks countermeasures (deny more than 10 connections every 10 minutes)" $IPT -A INPUT -p tcp --dport 22 -m recent --update --seconds 600 --hitcount 10 --rttl --name SSH -j LOGDROP echo "Allowing incoming traffic for HTTP, SMTP, NTP, PgSQL and SolR" $IPT -A INPUT -p tcp --dport 25 -i eth0 -j ACCEPT $IPT -A INPUT -p tcp --dport 80 -i eth0 -j ACCEPT $IPT -A INPUT -p udp --dport 123 -i eth0 -j ACCEPT $IPT -A INPUT -p tcp --dport 5433 -i eth0.2654 -s 172.16.0.2 -j ACCEPT $IPT -A INPUT -p udp --dport 5433 -i eth0.2654 -s 172.16.0.2 -j ACCEPT $IPT -A INPUT -p tcp --dport 8983 -i eth0.2654 -s 172.16.0.2 -j ACCEPT $IPT -A INPUT -p udp --dport 8983 -i eth0.2654 -s 172.16.0.2 -j ACCEPT echo "Allowing outgoing traffic for ICMP, SSH, whois, SMTP, DNS, HTTP, PgSQL and SolR" $IPT -A OUTPUT -p tcp --dport 22 -j ACCEPT $IPT -A OUTPUT -p tcp --dport 25 -o eth0 -j ACCEPT $IPT -A OUTPUT -p tcp --dport 43 -o eth0 -j ACCEPT $IPT -A OUTPUT -p tcp --dport 53 -o eth0 -j ACCEPT $IPT -A OUTPUT -p udp --dport 53 -o eth0 -j ACCEPT $IPT -A OUTPUT -p tcp --dport 80 -o eth0 -j ACCEPT $IPT -A OUTPUT -p udp --dport 80 -o eth0 -j ACCEPT #$IPT -A OUTPUT -p tcp --dport 5433 -o eth0 -d 176.31.236.101 -j ACCEPT #$IPT -A OUTPUT -p udp --dport 5433 -o eth0 -d 176.31.236.101 -j ACCEPT #$IPT -A OUTPUT -p tcp --dport 8983 -o eth0 -d 176.31.236.101 -j ACCEPT #$IPT -A OUTPUT -p udp --dport 8983 -o eth0 -d 176.31.236.101 -j ACCEPT $IPT -A OUTPUT -p tcp --sport 5433 -o eth0.2654 -j ACCEPT $IPT -A OUTPUT -p udp --sport 5433 -o eth0.2654 -j ACCEPT $IPT -A OUTPUT -p tcp --sport 8983 -o eth0.2654 -j ACCEPT $IPT -A OUTPUT -p udp --sport 8983 -o eth0.2654 -j ACCEPT $IPT -A OUTPUT -p icmp -j ACCEPT echo "Allowing outgoing FTP backup" $IPT -A OUTPUT -p tcp --dport 20:21 -o eth0 -d 91.121.190.78 -j ACCEPT echo "Dropping and logging everything else" $IPT -A INPUT -s 0/0 -j LOGDROP $IPT -A OUTPUT -j LOGDROP $IPT -A FORWARD -j LOGDROP echo "Firewall loaded." echo "Maintaining new rules for 3 minutes for tests" sleep 180 $IPT -nvL echo "Clearing firewall rules" $IPT -F $IPT -Z $IPT -t nat -F $IPT -t nat -Z $IPT -t mangle -F $IPT -t mangle -Z $IPT -X $IPT -P INPUT ACCEPT $IPT -P OUTPUT ACCEPT $IPT -P FORWARD ACCEPT When I launch this script (I only have a SSH access), the shell displays every message up to Maintaining new rules for 3 minutes for tests, the server is unresponsive during the 3 minutes delay and then resume normal operations. The only solution I found until now was to set $IPT -P INPUT ACCEPT and $IPT -I INPUT -i eth0 -j ACCEPT, but this configuration does not protect me of any attack, which is a great shame for a firewall. I suspect that the error comes from my script and not from iptables, but I don't understand what's wrong with my script. Could some do-gooder explain me my error, please? EDIT: here comes the result of iptables -nvL with the "accept all input" ($IPT -P INPUT ACCEPT and $IPT -I INPUT -i eth0 -j ACCEPT) solution : Chain INPUT (policy ACCEPT 0 packets, 0 bytes) pkts bytes target prot opt in out source destination 1 52 ACCEPT all -- eth0 * 0.0.0.0/0 0.0.0.0/0 0 0 ACCEPT all -- lo * 0.0.0.0/0 0.0.0.0/0 0 0 LOGDROP tcp -- eth0 * 0.0.0.0/0 0.0.0.0/0 tcp flags:0x17/0x02 limit: avg 100/sec burst 200 0 0 ACCEPT tcp -- * * 0.0.0.0/0 0.0.0.0/0 tcp dpt:22 state NEW recent: SET name: SSH side: source 0 0 LOGDROP tcp -- * * 0.0.0.0/0 0.0.0.0/0 tcp dpt:22 recent: UPDATE seconds: 600 hit_count: 10 TTL-Match name: SSH side: source 0 0 ACCEPT tcp -- eth0 * 0.0.0.0/0 0.0.0.0/0 tcp dpt:25 0 0 ACCEPT tcp -- eth0 * 0.0.0.0/0 0.0.0.0/0 tcp dpt:80 0 0 ACCEPT udp -- eth0 * 0.0.0.0/0 0.0.0.0/0 udp dpt:123 0 0 ACCEPT tcp -- eth0.2654 * 172.16.0.2 0.0.0.0/0 tcp dpt:5433 0 0 ACCEPT udp -- eth0.2654 * 172.16.0.2 0.0.0.0/0 udp dpt:5433 0 0 ACCEPT tcp -- eth0.2654 * 172.16.0.2 0.0.0.0/0 tcp dpt:8983 0 0 ACCEPT udp -- eth0.2654 * 172.16.0.2 0.0.0.0/0 udp dpt:8983 0 0 LOGDROP all -- * * 0.0.0.0/0 0.0.0.0/0 Chain FORWARD (policy ACCEPT 0 packets, 0 bytes) pkts bytes target prot opt in out source destination 0 0 LOGDROP all -- * * 0.0.0.0/0 0.0.0.0/0 Chain OUTPUT (policy ACCEPT 0 packets, 0 bytes) pkts bytes target prot opt in out source destination 0 0 ACCEPT all -- * lo 0.0.0.0/0 0.0.0.0/0 2 728 ACCEPT all -- * * 0.0.0.0/0 0.0.0.0/0 state RELATED,ESTABLISHED 0 0 ACCEPT tcp -- * * 0.0.0.0/0 0.0.0.0/0 tcp dpt:22 0 0 ACCEPT tcp -- * eth0 0.0.0.0/0 0.0.0.0/0 tcp dpt:25 0 0 ACCEPT tcp -- * eth0 0.0.0.0/0 0.0.0.0/0 tcp dpt:43 0 0 ACCEPT tcp -- * eth0 0.0.0.0/0 0.0.0.0/0 tcp dpt:53 0 0 ACCEPT udp -- * eth0 0.0.0.0/0 0.0.0.0/0 udp dpt:53 0 0 ACCEPT tcp -- * eth0 0.0.0.0/0 0.0.0.0/0 tcp dpt:80 0 0 ACCEPT udp -- * eth0 0.0.0.0/0 0.0.0.0/0 udp dpt:80 0 0 ACCEPT tcp -- * eth0.2654 0.0.0.0/0 0.0.0.0/0 tcp spt:5433 0 0 ACCEPT udp -- * eth0.2654 0.0.0.0/0 0.0.0.0/0 udp spt:5433 0 0 ACCEPT tcp -- * eth0.2654 0.0.0.0/0 0.0.0.0/0 tcp spt:8983 0 0 ACCEPT udp -- * eth0.2654 0.0.0.0/0 0.0.0.0/0 udp spt:8983 0 0 ACCEPT icmp -- * * 0.0.0.0/0 0.0.0.0/0 0 0 ACCEPT tcp -- * eth0 0.0.0.0/0 91.121.190.78 tcp dpts:20:21 0 0 LOGDROP all -- * * 0.0.0.0/0 0.0.0.0/0 Chain LOGDROP (5 references) pkts bytes target prot opt in out source destination 0 0 LOG all -- * * 0.0.0.0/0 0.0.0.0/0 limit: avg 5/min burst 5 LOG flags 0 level 7 prefix `iptables rejected: ' 0 0 DROP all -- * * 0.0.0.0/0 0.0.0.0/0 EDIT #2 : I modified my script (policy ACCEPT, defining authorized incoming packets then logging and dropping everything else) to write iptables -nvL results to a file and to allow only 10 ICMP requests per second, logging and dropping everything else. The result proved unexpected : while the server was unavailable to SSH connections, even already established, I ping-flooded it from another server, and the ping rate was restricted to 10 requests per second. During this test, I also tried to open new SSH connections, which remained unanswered until the script flushed rules. Here comes the iptables stats written after these tests : Chain INPUT (policy ACCEPT 0 packets, 0 bytes) pkts bytes target prot opt in out source destination 600 35520 ACCEPT all -- lo * 0.0.0.0/0 0.0.0.0/0 6 360 LOGDROP tcp -- eth0 * 0.0.0.0/0 0.0.0.0/0 tcp flags:0x17/0x02 limit: avg 100/sec burst 200 0 0 LOGDROP tcp -- eth0 * 0.0.0.0/0 0.0.0.0/0 tcp dpt:80 STRING match "w00tw00t.at.ISC.SANS." ALGO name bm TO 65535 0 0 LOGDROP tcp -- * * 0.0.0.0/0 0.0.0.0/0 tcp dpt:80 STRING match "Host: anoticiapb.com.br" ALGO name bm TO 65535 0 0 LOGDROP tcp -- * * 0.0.0.0/0 0.0.0.0/0 tcp dpt:80 STRING match "Host: www.anoticiapb.com.br" ALGO name bm TO 65535 105 8820 ACCEPT icmp -- * * 0.0.0.0/0 0.0.0.0/0 limit: avg 10/sec burst 5 830 69720 LOGDROP icmp -- * * 0.0.0.0/0 0.0.0.0/0 0 0 ACCEPT tcp -- * * 0.0.0.0/0 0.0.0.0/0 tcp dpt:22 state NEW recent: SET name: SSH side: source 0 0 LOGDROP tcp -- * * 0.0.0.0/0 0.0.0.0/0 tcp dpt:22 recent: UPDATE seconds: 600 hit_count: 10 TTL-Match name: SSH side: source 0 0 ACCEPT tcp -- eth0 * 0.0.0.0/0 0.0.0.0/0 tcp dpt:25 0 0 ACCEPT tcp -- eth0 * 0.0.0.0/0 0.0.0.0/0 tcp dpt:80 0 0 ACCEPT udp -- eth0 * 0.0.0.0/0 0.0.0.0/0 udp dpt:80 0 0 ACCEPT udp -- eth0 * 0.0.0.0/0 0.0.0.0/0 udp dpt:123 0 0 ACCEPT tcp -- eth0 * 0.0.0.0/0 0.0.0.0/0 tcp dpt:443 0 0 ACCEPT tcp -- eth0.2654 * 172.16.0.1 0.0.0.0/0 tcp spt:5433 0 0 ACCEPT udp -- eth0.2654 * 172.16.0.1 0.0.0.0/0 udp spt:5433 0 0 ACCEPT tcp -- eth0.2654 * 172.16.0.1 0.0.0.0/0 tcp spt:8983 0 0 ACCEPT udp -- eth0.2654 * 172.16.0.1 0.0.0.0/0 udp spt:8983 16 1684 LOGDROP all -- * * 0.0.0.0/0 0.0.0.0/0 Chain FORWARD (policy ACCEPT 0 packets, 0 bytes) pkts bytes target prot opt in out source destination 0 0 LOGDROP all -- * * 0.0.0.0/0 0.0.0.0/0 Chain OUTPUT (policy ACCEPT 0 packets, 0 bytes) pkts bytes target prot opt in out source destination 600 35520 ACCEPT all -- * lo 0.0.0.0/0 0.0.0.0/0 0 0 LOGDROP tcp -- * eth0 0.0.0.0/0 0.0.0.0/0 tcp dpt:80 owner UID match 33 0 0 LOGDROP udp -- * eth0 0.0.0.0/0 0.0.0.0/0 udp dpt:80 owner UID match 33 116 11136 ACCEPT all -- * * 0.0.0.0/0 0.0.0.0/0 state RELATED,ESTABLISHED 0 0 ACCEPT tcp -- * * 0.0.0.0/0 0.0.0.0/0 tcp dpt:22 0 0 ACCEPT tcp -- * eth0 0.0.0.0/0 0.0.0.0/0 tcp dpt:25 0 0 ACCEPT tcp -- * eth0 0.0.0.0/0 0.0.0.0/0 tcp dpt:53 0 0 ACCEPT udp -- * eth0 0.0.0.0/0 0.0.0.0/0 udp dpt:53 0 0 ACCEPT tcp -- * eth0 0.0.0.0/0 0.0.0.0/0 tcp dpt:80 0 0 ACCEPT udp -- * eth0 0.0.0.0/0 0.0.0.0/0 udp dpt:80 0 0 ACCEPT tcp -- * eth0.2654 0.0.0.0/0 0.0.0.0/0 tcp dpt:5433 0 0 ACCEPT udp -- * eth0.2654 0.0.0.0/0 0.0.0.0/0 udp dpt:5433 0 0 ACCEPT tcp -- * eth0.2654 0.0.0.0/0 0.0.0.0/0 tcp dpt:8983 0 0 ACCEPT udp -- * eth0.2654 0.0.0.0/0 0.0.0.0/0 udp dpt:8983 0 0 ACCEPT icmp -- * * 0.0.0.0/0 0.0.0.0/0 0 0 ACCEPT tcp -- * eth0 0.0.0.0/0 0.0.0.0/0 tcp dpt:43 0 0 ACCEPT tcp -- * eth0 0.0.0.0/0 91.121.190.18 tcp dpts:20:21 7 1249 LOGDROP all -- * * 0.0.0.0/0 0.0.0.0/0 Chain LOGDROP (11 references) pkts bytes target prot opt in out source destination 35 3156 LOG all -- * * 0.0.0.0/0 0.0.0.0/0 limit: avg 1/sec burst 5 LOG flags 0 level 7 prefix `iptables rejected: ' 859 73013 DROP all -- * * 0.0.0.0/0 0.0.0.0/0 Here comes the log content added during this test : Mar 28 09:52:51 localhost kernel: iptables rejected: IN=eth0 OUT= MAC=00:25:90:54:d7:88:10:8c:cf:28:39:80:08:00 SRC=194.51.74.245 DST=176.31.238.3 LEN=52 TOS=0x00 PREC=0x00 TTL=51 ID=55666 DF PROTO=TCP SPT=57504 DPT=22 WINDOW=501 RES=0x00 ACK URGP=0 Mar 28 09:52:51 localhost kernel: iptables rejected: IN=eth0 OUT= MAC=00:25:90:54:d7:88:10:8c:cf:28:39:80:08:00 SRC=194.51.74.245 DST=176.31.238.3 LEN=52 TOS=0x00 PREC=0x00 TTL=51 ID=55667 DF PROTO=TCP SPT=57504 DPT=22 WINDOW=501 RES=0x00 ACK URGP=0 Mar 28 09:52:51 localhost kernel: iptables rejected: IN=eth0 OUT= MAC=00:25:90:54:d7:88:10:8c:cf:28:39:80:08:00 SRC=194.51.74.245 DST=176.31.238.3 LEN=64 TOS=0x00 PREC=0x00 TTL=51 ID=55668 DF PROTO=TCP SPT=57504 DPT=22 WINDOW=501 RES=0x00 ACK URGP=0 Mar 28 09:52:51 localhost kernel: iptables rejected: IN=eth0 OUT= MAC=00:25:90:54:d7:88:10:8c:cf:28:39:80:08:00 SRC=194.51.74.245 DST=176.31.238.3 LEN=64 TOS=0x00 PREC=0x00 TTL=51 ID=55669 DF PROTO=TCP SPT=57504 DPT=22 WINDOW=501 RES=0x00 ACK URGP=0 Mar 28 09:52:52 localhost kernel: iptables rejected: IN=eth0 OUT= MAC=00:25:90:54:d7:88:10:8c:cf:28:39:80:08:00 SRC=194.51.74.245 DST=176.31.238.3 LEN=64 TOS=0x00 PREC=0x00 TTL=51 ID=55670 DF PROTO=TCP SPT=57504 DPT=22 WINDOW=501 RES=0x00 ACK URGP=0 Mar 28 09:52:54 localhost kernel: iptables rejected: IN=eth0 OUT= MAC=00:25:90:54:d7:88:10:8c:cf:28:39:80:08:00 SRC=194.51.74.245 DST=176.31.238.3 LEN=64 TOS=0x00 PREC=0x00 TTL=51 ID=55671 DF PROTO=TCP SPT=57504 DPT=22 WINDOW=501 RES=0x00 ACK URGP=0 Mar 28 09:52:58 localhost kernel: iptables rejected: IN=eth0 OUT= MAC=00:25:90:54:d7:88:10:8c:cf:28:39:80:08:00 SRC=194.51.74.245 DST=176.31.238.3 LEN=64 TOS=0x00 PREC=0x00 TTL=51 ID=55672 DF PROTO=TCP SPT=57504 DPT=22 WINDOW=501 RES=0x00 ACK URGP=0 Mar 28 09:52:59 localhost kernel: iptables rejected: IN=eth0 OUT= MAC=00:25:90:54:d7:88:10:8c:cf:28:39:80:08:00 SRC=176.31.236.101 DST=176.31.238.3 LEN=84 TOS=0x00 PREC=0x00 TTL=63 ID=0 DF PROTO=ICMP TYPE=8 CODE=0 ID=7430 SEQ=6 Mar 28 09:52:59 localhost kernel: iptables rejected: IN=eth0 OUT= MAC=00:25:90:54:d7:88:10:8c:cf:28:39:80:08:00 SRC=176.31.236.101 DST=176.31.238.3 LEN=84 TOS=0x00 PREC=0x00 TTL=63 ID=0 DF PROTO=ICMP TYPE=8 CODE=0 ID=7430 SEQ=7 Mar 28 09:52:59 localhost kernel: iptables rejected: IN=eth0 OUT= MAC=00:25:90:54:d7:88:10:8c:cf:28:39:80:08:00 SRC=176.31.236.101 DST=176.31.238.3 LEN=84 TOS=0x00 PREC=0x00 TTL=63 ID=0 DF PROTO=ICMP TYPE=8 CODE=0 ID=7430 SEQ=8 Mar 28 09:52:59 localhost kernel: iptables rejected: IN=eth0 OUT= MAC=00:25:90:54:d7:88:10:8c:cf:28:39:80:08:00 SRC=176.31.236.101 DST=176.31.238.3 LEN=84 TOS=0x00 PREC=0x00 TTL=63 ID=0 DF PROTO=ICMP TYPE=8 CODE=0 ID=7430 SEQ=9 Mar 28 09:52:59 localhost kernel: iptables rejected: IN=eth0 OUT= MAC=00:25:90:54:d7:88:10:8c:cf:28:39:80:08:00 SRC=176.31.236.101 DST=176.31.238.3 LEN=84 TOS=0x00 PREC=0x00 TTL=63 ID=0 DF PROTO=ICMP TYPE=8 CODE=0 ID=7430 SEQ=59 Mar 28 09:53:00 localhost kernel: iptables rejected: IN=eth0 OUT= MAC=00:25:90:54:d7:88:10:8c:cf:28:39:80:08:00 SRC=176.31.236.101 DST=176.31.238.3 LEN=84 TOS=0x00 PREC=0x00 TTL=63 ID=0 DF PROTO=ICMP TYPE=8 CODE=0 ID=7430 SEQ=152 Mar 28 09:53:01 localhost kernel: iptables rejected: IN=eth0 OUT= MAC=00:25:90:54:d7:88:10:8c:cf:28:39:80:08:00 SRC=176.31.236.101 DST=176.31.238.3 LEN=84 TOS=0x00 PREC=0x00 TTL=63 ID=0 DF PROTO=ICMP TYPE=8 CODE=0 ID=7430 SEQ=246 Mar 28 09:53:02 localhost kernel: iptables rejected: IN=eth0 OUT= MAC=00:25:90:54:d7:88:10:8c:cf:28:39:80:08:00 SRC=176.31.236.101 DST=176.31.238.3 LEN=84 TOS=0x00 PREC=0x00 TTL=63 ID=0 DF PROTO=ICMP TYPE=8 CODE=0 ID=7430 SEQ=339 Mar 28 09:53:03 localhost kernel: iptables rejected: IN=eth0 OUT= MAC=00:25:90:54:d7:88:10:8c:cf:28:39:80:08:00 SRC=176.31.236.101 DST=176.31.238.3 LEN=84 TOS=0x00 PREC=0x00 TTL=63 ID=0 DF PROTO=ICMP TYPE=8 CODE=0 ID=7430 SEQ=432 Mar 28 09:53:04 localhost kernel: iptables rejected: IN=eth0 OUT= MAC=00:25:90:54:d7:88:10:8c:cf:28:39:80:08:00 SRC=176.31.236.101 DST=176.31.238.3 LEN=84 TOS=0x00 PREC=0x00 TTL=63 ID=0 DF PROTO=ICMP TYPE=8 CODE=0 ID=7430 SEQ=524 Mar 28 09:53:05 localhost kernel: iptables rejected: IN=eth0 OUT= MAC=00:25:90:54:d7:88:10:8c:cf:28:39:80:08:00 SRC=176.31.236.101 DST=176.31.238.3 LEN=84 TOS=0x00 PREC=0x00 TTL=63 ID=0 DF PROTO=ICMP TYPE=8 CODE=0 ID=7430 SEQ=617 Mar 28 09:53:06 localhost kernel: iptables rejected: IN=eth0 OUT= MAC=00:25:90:54:d7:88:10:8c:cf:28:39:80:08:00 SRC=176.31.236.101 DST=176.31.238.3 LEN=84 TOS=0x00 PREC=0x00 TTL=63 ID=0 DF PROTO=ICMP TYPE=8 CODE=0 ID=7430 SEQ=711 Mar 28 09:53:07 localhost kernel: iptables rejected: IN=eth0 OUT= MAC=00:25:90:54:d7:88:10:8c:cf:28:39:80:08:00 SRC=176.31.236.101 DST=176.31.238.3 LEN=84 TOS=0x00 PREC=0x00 TTL=63 ID=0 DF PROTO=ICMP TYPE=8 CODE=0 ID=7430 SEQ=804 Mar 28 09:53:08 localhost kernel: iptables rejected: IN=eth0 OUT= MAC=00:25:90:54:d7:88:10:8c:cf:28:39:80:08:00 SRC=176.31.236.101 DST=176.31.238.3 LEN=84 TOS=0x00 PREC=0x00 TTL=63 ID=0 DF PROTO=ICMP TYPE=8 CODE=0 ID=7430 SEQ=897 Mar 28 09:53:16 localhost kernel: iptables rejected: IN=eth0 OUT= MAC=00:25:90:54:d7:88:c0:62:6b:e3:5c:80:08:00 SRC=194.51.74.245 DST=176.31.238.3 LEN=60 TOS=0x00 PREC=0x00 TTL=51 ID=61402 DF PROTO=TCP SPT=57637 DPT=22 WINDOW=5840 RES=0x00 SYN URGP=0 Mar 28 09:53:19 localhost kernel: iptables rejected: IN=eth0 OUT= MAC=00:25:90:54:d7:88:c0:62:6b:e3:5c:80:08:00 SRC=194.51.74.245 DST=176.31.238.3 LEN=60 TOS=0x00 PREC=0x00 TTL=51 ID=61403 DF PROTO=TCP SPT=57637 DPT=22 WINDOW=5840 RES=0x00 SYN URGP=0 Mar 28 09:53:21 localhost kernel: iptables rejected: IN=eth0 OUT= MAC=00:25:90:54:d7:88:10:8c:cf:28:39:80:08:00 SRC=194.51.74.245 DST=176.31.238.3 LEN=64 TOS=0x00 PREC=0x00 TTL=51 ID=55674 DF PROTO=TCP SPT=57504 DPT=22 WINDOW=501 RES=0x00 ACK URGP=0 Mar 28 09:53:25 localhost kernel: iptables rejected: IN=eth0 OUT= MAC=00:25:90:54:d7:88:c0:62:6b:e3:5c:80:08:00 SRC=194.51.74.245 DST=176.31.238.3 LEN=60 TOS=0x00 PREC=0x00 TTL=51 ID=61404 DF PROTO=TCP SPT=57637 DPT=22 WINDOW=5840 RES=0x00 SYN URGP=0 Mar 28 09:53:37 localhost kernel: iptables rejected: IN=eth0 OUT= MAC=00:25:90:54:d7:88:10:8c:cf:28:39:80:08:00 SRC=194.51.74.245 DST=176.31.238.3 LEN=116 TOS=0x00 PREC=0x00 TTL=51 ID=55675 DF PROTO=TCP SPT=57504 DPT=22 WINDOW=501 RES=0x00 ACK PSH URGP=0 Mar 28 09:53:37 localhost kernel: iptables rejected: IN=eth0 OUT= MAC=00:25:90:54:d7:88:10:8c:cf:28:39:80:08:00 SRC=194.51.74.245 DST=176.31.238.3 LEN=116 TOS=0x00 PREC=0x00 TTL=51 ID=55676 DF PROTO=TCP SPT=57504 DPT=22 WINDOW=501 RES=0x00 ACK PSH URGP=0 Mar 28 09:53:37 localhost kernel: iptables rejected: IN=eth0 OUT= MAC=00:25:90:54:d7:88:10:8c:cf:28:39:80:08:00 SRC=194.51.74.245 DST=176.31.238.3 LEN=180 TOS=0x00 PREC=0x00 TTL=51 ID=55677 DF PROTO=TCP SPT=57504 DPT=22 WINDOW=501 RES=0x00 ACK PSH URGP=0 Mar 28 09:53:38 localhost kernel: iptables rejected: IN=eth0 OUT= MAC=00:25:90:54:d7:88:10:8c:cf:28:39:80:08:00 SRC=194.51.74.245 DST=176.31.238.3 LEN=180 TOS=0x00 PREC=0x00 TTL=51 ID=55678 DF PROTO=TCP SPT=57504 DPT=22 WINDOW=501 RES=0x00 ACK PSH URGP=0 Mar 28 09:53:39 localhost kernel: iptables rejected: IN=eth0 OUT= MAC=00:25:90:54:d7:88:10:8c:cf:28:39:80:08:00 SRC=194.51.74.245 DST=176.31.238.3 LEN=180 TOS=0x00 PREC=0x00 TTL=51 ID=55679 DF PROTO=TCP SPT=57504 DPT=22 WINDOW=501 RES=0x00 ACK PSH URGP=0 Mar 28 09:53:39 localhost kernel: iptables rejected: IN=eth0 OUT= MAC=00:25:90:54:d7:88:c0:62:6b:e3:5c:80:08:00 SRC=194.51.74.245 DST=176.31.238.3 LEN=60 TOS=0x00 PREC=0x00 TTL=51 ID=5055 DF PROTO=TCP SPT=57638 DPT=22 WINDOW=5840 RES=0x00 SYN URGP=0 Mar 28 09:53:41 localhost kernel: iptables rejected: IN=eth0 OUT= MAC=00:25:90:54:d7:88:10:8c:cf:28:39:80:08:00 SRC=194.51.74.245 DST=176.31.238.3 LEN=180 TOS=0x00 PREC=0x00 TTL=51 ID=55680 DF PROTO=TCP SPT=57504 DPT=22 WINDOW=501 RES=0x00 ACK PSH URGP=0 Mar 28 09:53:42 localhost kernel: iptables rejected: IN=eth0 OUT= MAC=00:25:90:54:d7:88:c0:62:6b:e3:5c:80:08:00 SRC=194.51.74.245 DST=176.31.238.3 LEN=60 TOS=0x00 PREC=0x00 TTL=51 ID=5056 DF PROTO=TCP SPT=57638 DPT=22 WINDOW=5840 RES=0x00 SYN URGP=0 Mar 28 09:53:45 localhost kernel: iptables rejected: IN=eth0 OUT= MAC=00:25:90:54:d7:88:10:8c:cf:28:39:80:08:00 SRC=194.51.74.245 DST=176.31.238.3 LEN=180 TOS=0x00 PREC=0x00 TTL=51 ID=55681 DF PROTO=TCP SPT=57504 DPT=22 WINDOW=501 RES=0x00 ACK PSH URGP=0 Mar 28 09:53:48 localhost kernel: iptables rejected: IN=eth0 OUT= MAC=00:25:90:54:d7:88:c0:62:6b:e3:5c:80:08:00 SRC=194.51.74.245 DST=176.31.238.3 LEN=60 TOS=0x00 PREC=0x00 TTL=51 ID=5057 DF PROTO=TCP SPT=57638 DPT=22 WINDOW=5840 RES=0x00 SYN URGP=0 If I correctly interpreted these results, they say that ICMP rules were correctly interpreted by iptables, but SSH rules were not. This does not make any sense... Does somebody understand where my error comes from? EDIT #3 : After some more tests, I found out that commenting the SYN flood countermeasure removes the problem. I continue researches in this way but, meanwhile, if somebody sees my anti SYN flood rule error...

    Read the article

  • Real server, Multiple IP Addresses, HyperV Virtual Server, How to partition IPs across real and Virtual NICs

    - by Steven_W
    This is a slightly difficult problem to explain without same basic background information - I'll try and refine the question later as necessary Originally, I have a single hosted server (Win 2008R2) with the following range of 8 IP addresses. - Single NIC - IP: x.x.128.72 -> x.x.128.79 - Subnet: x.x.255.192 - GW: x.x.128.65 After installing Hyper-V and setting up a single virtual server on the same box, I then wanted to assign one of the IP addresses to the virtual server, leaving everything else running normally. -- Firstly, I tried using the "External" network, but (even after setting IPs on the "Virtual Adapter" similar to Here but struggled to get networking running at all. I needed to keep the server running (otherwise I would have spent more time pursuing this approach) Q1 ... Was this a sensible thing to do ? Should I have carried on down this route ? -- I then decided to try different approach - Set the HyperV network to "Internal" (visible to Management OS) - Physical NIC - IP: x.x.128.72 -> x.x.128.75 - Subnet: x.x.255.192 - GW: x.x.128.65 - Virtual NIC - IP: x.x.128.78 - Subnet: x.x.255.252 - GW: x.x.128.72 ... { The same as the IP of the physical NIC ) - Virtual OS-NIC - IP: x.x.128.77 - Subnet: x.x.255.252 - GW: x.x.128.78 ... { The same as the IP of the host virtual-NIC ) -- Surprisingly enough, this approach actually worked, and I was able to connect from all the following: - Internet to/from physical NIC (x.x.128.72) - physical NIC (x.x.128.72) to virtual-OS-NIC (x.x.128.77) e.g. testing via ping + FTP - Internet to/from virtual-OS-NIC (x.x.128.72) -- The problem I have is that this approach seems to only last for a short while (a few hours). After this time, it seems that I lose the ability to connect from Virtual-OS-NIC to/from the internet (but I can still connect from the host-OS to the virtual-OS and from the host-OS to the internet) I have re-tested this a couple of times with the same results ... I leave the server on for a few hours (e.g. overnight), and when I come back in the morning, the Virtual-OS loses the ability to route to the internet -- I'm not quite sure what to look at next (or whether I'm going about this completely the wrong way ) One "possible relevant item" is that the host-OS is also running RRAS (Routing and Remote Access), but this is only to run a simple VPN -- Q2 - Wheat should I be looking at next ? (Any good references / recommendations of what to try) Would appreciate any thoughts or comments (even if you tell me I'm going about this the wrong way)

    Read the article

  • Automating the choice between JPEG and PNG with a script

    - by MHC
    Choosing the right format to save your images in is crucial for preserving image quality and reducing artifacts. Different formats follow different compression methods and come with their own set of advantages and disadvantages. JPG, for instance is suited for real life photographs that are rich in color gradients. The lossless PNG, on the other hand, is far superior when it comes to schematic figures: Picking the right format can be a chore when working with a large number of files. That's why I would love to find a way to automate it. A little bit of background on my particular use case: I am working on a number of handouts for a series of lectures at my unversity. The handouts are rich in figures, which I have to extract from PDF-formatted slides. Extracting these images gives me lossless PNGs, which are needlessly large at times. Converting these particular files to JPEG can reduce their size to up to less than 20% of their original file size, while maintaining the same quality. This is important as working with hundreds of large images in word processors is pretty crash-prone. Batch converting all extracted PNGs to JPEGs is not an option I am willing to follow, as many if not most images are better suited to be formatted as PNGs. Converting these would result in insignificant size reductions and sometimes even increases in filesize - that's at least what my test runs showed. What we can take from this is that file size after compression can serve as an indicator on what format is suited best for a particular image. It's not a particularly accurate predictor, but works well enough. So why not use it in form of a script: I included inotifywait because I would prefer for the script be executed automatically as soon as I drag an extracted image into a folder. This is a simpler version of the script that I've been using for the last couple of weeks: #!/bin/bash inotifywait -m --format "%w%f" --exclude '.jpg' -r -e create -e moved_to --fromfile '/home/MHC/.scripts/Workflow/Conversion/include_inotifywait' | while read file; do mogrify -format jpg -quality 92 "$file" done The advanced version of the script would have to be able to handle spaces in file names and directory names preserve the original file names flatten PNG images if an alpha value is set compare the file size between the temporary converted image and its original determine if the difference is greater than a given precentage act accordingly The actual conversion could be done with imagemagick tools: convert -quality 92 -flatten -background white file.png file.jpg Unfortunately, my bash skills aren't even close to advanced enough to convert the scheme above into an actual script, but I am sure many of you can. My reputation points on here are pretty low, but I will gladly award the most helpful answer with the highest bounty I can set. References: http://www.formortals.com/introducing-cnb-imageguide/, http://www.turnkeylinux.org/blog/png-vs-jpg Edit: Also see my comments below for some more information on why I think this script would be the best solution to the problem I am facing.

    Read the article

  • Rails 3 shows 404 error instead of index.html (nginx + unicorn)

    - by Miko
    I have an index.html in public/ that should be loading by default but instead I get a 404 error when I try to access http://example.com/ The page you were looking for doesn't exist. You may have mistyped the address or the page may have moved. This has something to do with nginx and unicorn which I am using to power Rails 3 When take unicorn out of the nginx configuration file, the problem goes away and index.html loads just fine. Here is my nginx configuration file: upstream unicorn { server unix:/tmp/.sock fail_timeout=0; } server { server_name example.com; root /www/example.com/current/public; index index.html; keepalive_timeout 5; location / { try_files $uri @unicorn; } location @unicorn { proxy_pass http://unicorn; proxy_set_header X-Forwarded-For $proxy_add_x_forwarded_for; proxy_set_header Host $host; proxy_redirect off; } } My config/routes.rb is pretty much empty: Advertise::Application.routes.draw do |map| resources :users end The index.html file is located in public/index.html and it loads fine if I request it directly: http://example.com/index.html To reiterate, when I remove all references to unicorn from the nginx conf, index.html loads without any problems, I have a hard time understanding why this occurs because nginx should be trying to load that file on its own by default. -- Here is the error stack from production.log: Started GET "/" for 68.107.80.21 at 2010-08-08 12:06:29 -0700 Processing by HomeController#index as HTML Completed in 1ms ActionView::MissingTemplate (Missing template home/index with {:handlers=>[:erb, :rjs, :builder, :rhtml, :rxml, :haml], :formats=>[:html], :locale=>[:en, :en]} in view paths "/www/example.com/releases/20100808170224/app/views", "/www/example.com/releases/20100808170224/vendor/plugins/paperclip/app/views", "/www/example.com/releases/20100808170224/vendor/plugins/haml/app/views"): /usr/local/rvm/gems/ruby-1.9.2-rc2/gems/actionpack-3.0.0.beta4/lib/action_view/paths.rb:14:in `find' /usr/local/rvm/gems/ruby-1.9.2-rc2/gems/actionpack-3.0.0.beta4/lib/action_view/lookup_context.rb:79:in `find' /usr/local/rvm/gems/ruby-1.9.2-rc2/gems/actionpack-3.0.0.beta4/lib/action_view/base.rb:186:in `find_template' /usr/local/rvm/gems/ruby-1.9.2-rc2/gems/actionpack-3.0.0.beta4/lib/action_view/render/rendering.rb:45:in `_determine_template' /usr/local/rvm/gems/ruby-1.9.2-rc2/gems/actionpack-3.0.0.beta4/lib/action_view/render/rendering.rb:23:in `render' /usr/local/rvm/gems/ruby-1.9.2-rc2/gems/haml-3.0.15/lib/haml/helpers/action_view_mods.rb:13:in `render_with_haml' etc... -- nginx error log for this virtualhost comes up empty: 2010/08/08 12:40:22 [info] 3118#0: *1 client 68.107.80.21 closed keepalive connection My guess is unicorn is intercepting the request to index.html before nginx gets to process it.

    Read the article

  • How to correctly partition usb flash drive and which filesystem to choose considering wear leveling?

    - by random1
    Two problems. First one: how to partition the flash drive? I shouldn't need to do this, but I'm no longer sure if my partition is properly aligned since I was forced to delete and create a new partition table after gparted complained when I tried to format the drive from FAT to ext4. The naive answer would be to say "just use default and everything is going to be alright". However if you read the following links you'll know things are not that simple: https://lwn.net/Articles/428584/ and http://linux-howto-guide.blogspot.com/2009/10/increase-usb-flash-drive-write-speed.html Then there is also the issue of cylinders, heads and sectors. Currently I get this: $sfdisk -l -uM /dev/sdd Disk /dev/sdd: 30147 cylinders, 64 heads, 32 sectors/track Warning: The partition table looks like it was made for C/H/S=*/255/63 (instead of 30147/64/32). For this listing I'll assume that geometry. Units = mebibytes of 1048576 bytes, blocks of 1024 bytes, counting from 0 Device Boot Start End MiB #blocks Id System /dev/sdd1 1 30146 30146 30869504 83 Linux $fdisk -l /dev/sdd Disk /dev/sdd: 31.6 GB, 31611420672 bytes 255 heads, 63 sectors/track, 3843 cylinders Units = cylinders of 16065 * 512 = 8225280 bytes Sector size (logical/physical): 512 bytes / 512 bytes I/O size (minimum/optimal): 512 bytes / 512 bytes Disk identifier: 0x00010c28 So from my current understanding I should align partitions at 4 MiB (currently it's at 1 MiB). But I still don't know how to set the heads and sectors properly for my device. Second problem: file system. From the benchmarks I saw ext4 provides the best performance, however there is the issue of wear leveling. How can I know that my Transcend JetFlash 700's microcontroller provides for wear leveling? Or will I just be killing my drive faster? I've seen a lot of posts on the web saying don't worry the newer drives already take care of that. But I've never seen a single piece of backed evidence of that and at some point people start mixing SSD with USB flash drives technology. The safe option would be to go for ext2, however a serious of tests that I performed showed horrible performance!!! These values are from a real scenario and not some synthetic test: 42 files: 3,429,415,284 bytes copied to flash drive original fat32: 15.1 MiB/s ext4 after new partition table: 10.2 MiB/s ext2 after new partition table: 1.9 MiB/s Please read the links that I posted above before answering. I would also be interested in answers backed up with some references because a lot is said and re-said but then it lacks facts. Thank you for the help.

    Read the article

  • How can I prevent an unintentional DDOS running ColdFusion 8 with IIS 6?

    - by Eric Belair
    We had an interesting outage today on one of our client's websites. Out of nowhere, the website was inaccessible. The website runs by itself on a dedicated physical Windows 2000 server (probably overkill, I know, but that's a discussion for a different day). After restarting IIS and ColdFusion Application Service, the problem came back several times. My initial thought was that it was a DNS issue, which happens occasionally - the last time it happened was after Hurricane Sandy when we our ISP was out, and we had to make some network config changes. But, it was not a DNS issue. My second thought was that it was a DDOS attack, but, there's very little reason anyone would want to take this site down. When we called our ISP, the operator on the other end noted that traffic was spiking significantly. As it turned out, the client had unintentionally caused a DDOS on the website, after they FTPed a very large video file, and then mass emailed a link to it. Hundreds of people clicked the link and brought the site to its knees. I am primarily a Website Programmer, but I often have to contribute to server administration at times. Sadly, I'm the resident ColdFusion and IIS expert, but I don't have a lot of experience with this issue. What are some basic steps that I can take to prevent this from happening in the future, since we cannot always control what files the client posts to the website. Here are some ideas I had, but I'm unsure of the impact: Limit the number of connections in IIS. Put media files on a separate server (like an Amazon site, etc.). File requests of this type currently behind a server-script (i.e. /www.site.com/viewFile.cfm?fileId=1424545, where the fileId references a file off the webroot) that logs requests, and pushes the file to the browser using CFCONTENT. I could edit this script to reject requests when they exceed a certain amount in a given time-frame (i.e. a 5MB can be accessed globally 10 times in an hour). This may cause some users frustration, but, if hundreds of users are attempting to view the file, the site is going to crash anyways, as it did today, which is way more frustrating, since there is no "pretty" message explaining why they can't get to the file. I'm open to any suggestions, as I'm continuing my research to report to the CTO with the best options, so that we can put a solution into effect. Thank you.

    Read the article

  • CentOS 6 - Make system aware of custom lib paths and missing base links

    - by Mike Purcell
    I am trying to compile libmemcached (1.0.7) on CentOS6, and keep getting the following warning: ... checking for event.h... no configure: WARNING: Unable to find libevent ... I manually compiled libevent (2.0.19) and built it using the following configure line: OPTIONS="--prefix=/usr/local/_custom/app/libevent" Everything compiled and installed fine, but I couldn't figure out how to make the system aware that the lib files are in the custom /usr/local/_custom/app/libevent/libdir. I stumbled upon an article and read that I can make the system aware of custom lib paths by adding a custom file to /etc/ld.so.conf.d/ directory: # /etc/ld.so.conf.d/customApp.conf /usr/local/_custom/app/libevent/lib Then I issued the ldconfig command and was able to confirm that libevent was included by issuing this command: ldconfig -p | ack -i libevent Seeing that libevent was now included in the ldconfig output, I figured I would be able to compile libmemcached and satisfy the aforementioned warning. Unfortunately it did not. So I took another look at the ldconfig output and noticed this: libevent_pthreads-2.0.so.5 (libc6,x86-64) => /usr/local/_custom/app/libevent/lib/libevent_pthreads-2.0.so.5 libevent_openssl-2.0.so.5 (libc6,x86-64) => /usr/local/_custom/app/libevent/lib/libevent_openssl-2.0.so.5 libevent_extra-2.0.so.5 (libc6,x86-64) => /usr/local/_custom/app/libevent/lib/libevent_extra-2.0.so.5 libevent_core-2.0.so.5 (libc6,x86-64) => /usr/local/_custom/app/libevent/lib/libevent_core-2.0.so.5 libevent-2.0.so.5 (libc6,x86-64) => /usr/local/_custom/app/libevent/lib/libevent-2.0.so.5 There are no references to the base links, for example, I would expect to see links to these (ls -la /usr/local/_custom/app/libevent/lib): libevent.so -> libevent-2.0.so.5.1.7 libevent_openssl.so -> libevent_openssl-2.0.so.5.1.7 libevent_core.so -> libevent_core-2.0.so.5.1.7 So either I am doing something wrong, or the system still does not know where to look to find libevent.so. -- Update #1 -- I wasn't able to get libmemcached to compile without the warning notice, even after trying to compile using the following configure command: ./configure --prefix=/usr/local/_custom/app/libmemcached CFLAGS="-I/usr/local/_custom/app/libevent/include" LDFLAGS="-L/usr/local/_custom/app/libevent/lib" I thought for sure this would work because I am directly passing the include and lib directories to the configure command. But it did not.

    Read the article

  • Email can't be sent to my domain

    - by Jack W-H
    Hi Folks, Basically I have my domain howcode.com bought at DomainMonster.com. I have set it all up to point to MediaTemple nameservers and everything works - mostly - fine. I have registered an email address [email protected]. The setup is, I presume, working correctly. I can successfully send emails with the account. And I presume I can receive them - but the problem is, nobody can send them to me. Emailing from a regular, non-Googlemail account appears to work fine but it never arrives in the inbox. But when you email from a GoogleMail address, an error message is instantly returned saying this: Delivery to the following recipient failed permanently: [email protected] Technical details of permanent failure: Google tried to deliver your message, but it was rejected by the recipient domain. We recommend contacting the other email provider for further information about the cause of this error. The error that the other server returned was: 550 550 relay not permitted (state 14). ----- Original message ----- Received: by 10.216.91.12 with SMTP id g12mr3673969wef.77.1271503997091; Sat, 17 Apr 2010 04:33:17 -0700 (PDT) Return-Path: Received: from [192.168.0.3] (client-81-98-94-79.cht-bng-014.adsl.virginmedia.net [81.98.94.79]) by mx.google.com with ESMTPS id x1sm29451927wbx.19.2010.04.17.04.33.15 (version=TLSv1/SSLv3 cipher=RC4-MD5); Sat, 17 Apr 2010 04:33:16 -0700 (PDT) From: Jack Webb-Heller Mime-Version: 1.0 (Apple Message framework v1078) Content-Type: multipart/alternative; boundary=Apple-Mail-7--1008464685 Subject: Re: Hi Date: Sat, 17 Apr 2010 12:33:14 +0100 In-Reply-To: <[email protected] To: Jack Webb-Heller References: <[email protected] Message-Id: <[email protected] X-Mailer: Apple Mail (2.1078) Does this work? On 17 Apr 2010, at 12:32, Jack Webb-Heller wrote: Hi I thought this may be something to do with my MX DNS settings. They are setup like so: MX name: howcode.com TTL: 43200 Type: MX Data: 10 mail.howcode.com. The A-Record for mail.howcode.com is setup like this: Name: mail.howcode.com TTL: 43200 Type: A Data: 205.186.187.129 Is this what's going wrong with the issue? Thanks very much Jack

    Read the article

  • How to stop a random ramp in FCGI Processes Killing the server

    - by Andy Main
    So got the below earlier to day... Around that time the logs show a ramp in processes(600) and associated memory (1.2g), cpu usage load average (80) untill the server gave out. Server had to be hard reset by host as there was no ssh or plesk panel access. Fast CGI is configured as below and is setup for one high use site. As I understand it FcgidMaxProcesses 20 should protect against what happen but has not. I've read many forums with differing answers and references to many different fcgi directives, but have found nothing conclusive. Any one got some definitive answers on how to stop this sort of server process ramping and subsequent server failure? If you need more info let me know. Cheers Andy  /var/log/apache2/error_log [Thu May 17 07:40:47 2012] [warn] mod_fcgid: process 17651 graceful kill fail, sending SIGKILL [Thu May 17 07:40:47 2012] [warn] mod_fcgid: process 17650 graceful kill fail, sending SIGKILL [Thu May 17 07:40:47 2012] [warn] mod_fcgid: process 17649 graceful kill fail, sending SIGKILL [Thu May 17 07:40:47 2012] [warn] mod_fcgid: process 17644 graceful kill fail, sending SIGKILL [Thu May 17 07:40:47 2012] [warn] mod_fcgid: process 17643 graceful kill fail, sending SIGKILL [Thu May 17 07:40:47 2012] [warn] mod_fcgid: process 17638 graceful kill fail, sending SIGKILL [Thu May 17 07:40:47 2012] [warn] mod_fcgid: process 17633 graceful kill fail, sending SIGKILL [Thu May 17 07:40:47 2012] [warn] mod_fcgid: process 17627 graceful kill fail, sending SIGKILL [Thu May 17 07:40:47 2012] [warn] mod_fcgid: process 17622 graceful kill fail, sending SIGKILL [Thu May 17 07:40:51 2012] [warn] mod_fcgid: process 17674 graceful kill fail, sending SIGKILL [Thu May 17 07:40:51 2012] [warn] mod_fcgid: process 17673 graceful kill fail, sending SIGKILL [Thu May 17 07:40:51 2012] [warn] mod_fcgid: process 17672 graceful kill fail, sending SIGKILL [Thu May 17 07:40:51 2012] [warn] mod_fcgid: process 17667 graceful kill fail, sending SIGKILL [Thu May 17 07:40:51 2012] [warn] mod_fcgid: process 17666 graceful kill fail, sending SIGKILL [Thu May 17 07:40:51 2012] [warn] mod_fcgid: process 17665 graceful kill fail, sending SIGKILL [Thu May 17 07:40:51 2012] [warn] mod_fcgid: process 17664 graceful kill fail, sending SIGKILL [Thu May 17 07:40:51 2012] [warn] mod_fcgid: process 17659 graceful kill fail, sending SIGKILL [Thu May 17 07:40:51 2012] [warn] mod_fcgid: process 17658 graceful kill fail, sending SIGKILL [Thu May 17 07:40:51 2012] [warn] mod_fcgid: process 17657 graceful kill fail, sending SIGKILL [Thu May 17 07:40:51 2012] [warn] mod_fcgid: process 17656 graceful kill fail, sending SIGKILL [Thu May 17 07:40:51 2012] [warn] mod_fcgid: process 17651 graceful kill fail, sending SIGKILL https://docs.google.com/a/thesugarrefinery.com/open?id=0B_XbpWChge0VRmFLWEZfR2VBb2M https://docs.google.com/a/thesugarrefinery.com/open?id=0B_XbpWChge0VWTcwZEhoV2Fqejg https://docs.google.com/a/thesugarrefinery.com/open?id=0B_XbpWChge0VUUtVWWFINHZjZ0U https://docs.google.com/a/thesugarrefinery.com/open?id=0B_XbpWChge0VZEVMclh6ZUdaOUE <IfModule mod_fcgid.c> <IfModule !mod_fastcgi.c> AddHandler fcgid-script fcg fcgi fpl </IfModule> FcgidIPCDir /var/lib/apache2/fcgid/sock FcgidProcessTableFile /var/lib/apache2/fcgid/shm FcgidIdleTimeout 40 FcgidProcessLifeTime 30 FcgidMaxProcesses 20 FcgidMaxProcessesPerClass 20 FcgidMinProcessesPerClass 0 FcgidConnectTimeout 30 FcgidIOTimeout 120 FcgidInitialEnv RAILS_ENV production FcgidIdleScanInterval 10 FcgidMaxRequestLen 1073741824 </IfModule>

    Read the article

  • Unable to use TweetDeck on Windows due to "Ooops, TweetDeck can't find your data" and "Sorry, Adobe

    - by Matt
    I'm running Adobe AIR 1.5.2 (latest) on Windows 7 (64-bit RTM) and downloaded TweetDeck 0.31.1 (latest). When I run TweetDeck I get the following errors: Ooops, TweetDeck can't find your data and Sorry, Adobe AIR has a problem running on this computer Other AIR applications install and run fine. I've uninstalled both TweetDeck and AIR and reinstalled. Following the uninstalls I've also removed all on-disk references to both TweetDeck and AIR, but no luck. UPDATE: Using Process Monitor I did a trace of Tweetdeck from the moment it launched until the first error occurred. I saw the following information in the output of the trace: 1 5:22:18.6522338 PM TweetDeck.exe 5580 CreateFile D:\ProgramData\Microsoft\Windows\Start Menu\Programs\rs\??\d:\Use\myusername\AppData\Roaming\Adobe\AIR\ELS\TweetDeckFast.F9107117265DB7542C1A806C8DB837742CE14C21.1\PrivateEncryptedDatak NAME INVALID Desired Access: Generic Write, Read Attributes, Disposition: OverwriteIf, Options: Synchronous IO Non-Alert, Non-Directory File, Attributes: N, ShareMode: Read, Write, AllocationSize: 0 In this trace output, Tweetdeck.exe is trying to create the file D:\ProgramData\Microsoft\Windows\Start Menu\Programs\rs\??\d:\Use\myusername\AppData\Roaming\Adobe\AIR\ELS\TweetDeckFast.F9107117265DB7542C1A806C8DB837742CE14C21.1\PrivateEncryptedDatak but the path specified is invalid. When looking at the path you can see that it is indeed an invalid path. First, there’s the “??” portion which doesn’t exist in the file system since the “?” is an invalid character in Windows/NTFS file systems. Additionally, looking at this path, it actually seems to be composed of two parts (is the "??" a delimiter?): Part 1: D:\ProgramData\Microsoft\Windows\Start Menu\Programs\rs\?? Part 2: d:\Use\myusername\AppData\Roaming\Adobe\AIR\ELS \TweetDeckFast.F9107117265DB7542C1A806C8DB837742CE14C21.1\PrivateEncryptedDatak (the problem here is that d:\Use... doesn’t even exist. What seems to be happening here is that Tweetdeck is looking for the user credentials (the “PrivateEncryptedDatak” file) but it’s looking in the wrong place, can’t find the file, and hence the error that Tweetdeck is giving (shown in the screenshot). I'm trying to determine how TweetDeck is getting this path. I searched the contents of all files on my hard disk hoping to find some TweetDeck or Adobe AIR configuration file containing this incorrect path, but I was unable to find anything. UPDATE: See Carl's comment regarding directory junctions and symbolic links under my accepted answer. This ended up being the problem. Edit by Gnoupi: People, the answer section is there to provide an actual ANSWER, not to say you have the same issue. It doesn't help anyone that you have the same problem. Eventually, if you think this is really worth mentioning, put it as a comment under the question. But simply, if what you want to add is not an answer to the question, then don't post it as an answer. This is not a forum, I recommend new users to read the FAQ: http://superuser.com/faq

    Read the article

  • Email Tracking - By IP

    - by disasm
    I am trying to track an Email I received to see where it originated from (harmful scam Email). Here is the full header: Received-SPF: pass (domain of aol.com designates 205.188.109.203 as permitted sender) ZSwgeW91ciBpbmZvcm1hdGlvbiBoYXMgYmVlbiBhY2NlcHRlZCBhbmQgYXBw cm92ZWQgYnkgb3VyIFJlY3J1aXRpbmcgRGVwYXJ0bWVudC4gWU9VUiBGSVJT VCBBU1NJR05NRU5UIERFVEFJTFMuIFdlc3Rlcm4gVW5pb24gTW9uZXkgVHJh bnNmZXIgaGFkIHJlcG9ydHMgYWJvdXQgbGFwc2VzIGluIHRoZSBzZXJ2aWNl cyBvZiBzb21lIG9mIHRoZWlyIG91dGxldHMgYW5kIHNvbWUgbwEwAQEBAQN0 ZXh0L3BsYWluAwMwAgN0ZXh0L2h0bWwDAzIx X-YMailISG: HeDj35sWLDvNbQaO2B72kIm7hhiubE2qUysyBFo2m3GE4wsk FOs6uaYWwBVUBbL_ubGEs5Gitm1b86QFPuQYdS3g5s2f_uY3dyHhXh7DcBIB Ad4mJBzeRozs5.0s6vbqhsIEYlaKI4EDrsocJEbDbTUiUq2UyxZ7Ery8Iqow _sBVN0msHJvcI09KwmaXUteV_qCL7qFlj7WNSmdMM.wVUm3pyiWzw1VUZlyD nwoEzdEImQdwmJqoTM1YE8XU6BE8IkmUkh1Z8XkfLtHqmAsPi1_.Msbi8ubk QD71BcTABjb7ixELg5NfomsyZKVN.9G.TnuISlX5umByjS701ITyQ2PUYXai hoVCWg37bKWNR9MAWdogUK8PIV3MWPv5gdglNAKuPdS5Z7.01J39UGyH7R60 aIiIWdAsQ7_3VQBgIi9Seg2YM2j1U5g9QtdcJxBe0.1oigmj7G2sC9.YXNGX 3abQ1EcWVlJLuSuBbQ4Flpbe_Y3_ssz8nZIK2YjKy0U8WWe77vfnxdEBsknf w2OA_PAzHtuAAuxETnAOU_MeMIssgRAtihKC_26Au1LnKYCGPGADFBLaLNHF 30itI.kBvUjdvUfqV11dnGe50kFVzBGDMJFm8mXvb5WtIKq6qU1ZZmUroCew EgXjVZ.JKbux2KQmHh2zZbIJO3nOmLGkzuRczYiWCUNBDtmUZE6imuIQ4P6S RjusSbMITf2fIL_xe.qFCnW563sOdc4u.uXLDx.lq30740l8lWkkLX6KaDMF k9TY0VQKsMynqa4vXKpkTVNdukAcGd0p2i3newxY4q_9eZLn9czsJimfpKNW SX1bqjs0iCQHb4FTydf1Zpa2b.6lIhdjVlIM8tiWhfGhlUeM267T3njEM6nz 0vxyjparR_G_s0VnIVhSeLw2F5KpAL226w2yA.WBcqoG2ROSa7fK.0ZYwy36 Qcmk8C.HKj8Fng1qFLtEfaI4F66rCEJi7h1d6EK0Jk4a_TJnBBub1VQVoU.s SJ2ehs8aDjDqJw27_Ia4vYekKhIU8Oak0vYSmMXhZ5IvJfGfOHYVy4ebkoQf IDE3lSfex1nHZqcMqq0agPOZUOdznSIGJVx4T8m6MGwrEouvL.grhT6KUJQ5 g8UX6DVTgj.8lHuTyOzj3A3NRwDFs2JqicprOMJRS4UWYX8eQ1y4j.4ora36 LnWYm7k1n6X0lDBW5ZdZlsLy7.0al0G1uCIAZwBNo7FnHr6q2mQNwgFaPkNO FOiykqFHu0khLO_cZw87MpDslZO_3lFbJGlnchSs81hkESSQsldUxqdNkIV. yWsS58p1uuwVNksp4NB.QW41wfBtY5FU.Q80g8KiOZIz0daou3GlzoahcHoQ GPgSa86GKtSo.ew2xEUKk6c.ffAT9RjqNh5fzyhBdzEYURxJBYgMlL5DQp2G yYIGhlIS5h9JzPFVkk2XhBoY2NgEAAfJfAfqKoMNNKIW.bEwbgNa9xtSzHNg YdmDfOSkYkAGZDqwa.uONguq5.jqtnWDnx3GDyuoVg-- X-Originating-IP: [205.188.109.203] Authentication-Results: mta1343.mail.bf1.yahoo.com from=aol.com; domainkeys=neutral (no sig); from=mx.aol.com; dkim=pass (ok) Received: from 127.0.0.1 (EHLO omr-d06.mx.aol.com) (205.188.109.203) by mta1343.mail.bf1.yahoo.com with SMTP; Fri, 18 Oct 2013 10:11:15 -0700 Received: from mtaomg-da05.r1000.mx.aol.com (mtaomg-da05.r1000.mx.aol.com [172.29.51.141]) by omr-d06.mx.aol.com (Outbound Mail Relay) with ESMTP id ACFAD7012DFB4 for <.confidential.; Fri, 18 Oct 2013 13:11:15 -0400 (EDT) Received: from core-mfb002c.r1000.mail.aol.com (core-mfb002.r1000.mail.aol.com [172.29.47.199]) by mtaomg-da05.r1000.mx.aol.com (OMAG/Core Interface) with ESMTP id 2ADB6E000089 for <.confidential.; Fri, 18 Oct 2013 13:11:15 -0400 (EDT) References: <[email protected] <[email protected] <[email protected] To: .confidential. Subject:.confidential. In-Reply-To: <[email protected] X-MB-Message-Source: WebUI MIME-Version: 1.0 From: .confidential. X-MB-Message-Type: User Content-Type: multipart/alternative; boundary="--------MB_8D09A3C2FD3105D_1338_33A66_webmail-d257.sysops.aol.com" X-Mailer: AOL Webmail 38109-STANDARD Received: from 66.199.226.81 by webmail-d257.sysops.aol.com (205.188.17.42) with HTTP (WebMailUI); Fri, 18 Oct 2013 13:11:15 -0400 Message-Id: <[email protected] X-Originating-IP: [66.199.226.81] Date: Fri, 18 Oct 2013 13:11:15 -0400 (EDT) x-aol-global-disposition: G DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=mx.aol.com; s=20121107; t=1382116275; bh=9TXLF90L8beaMnjNzoKDwcv3Eq06jiZGN40YTBw2YOI=; h=From:To:Subject:Message-Id:Date:MIME-Version:Content-Type; b=xHVjoH5AccrOpPZoZZW+b41uJ7nzHDrryGsO6WzvtBOFGWX3xJMO3RB1ILFlJAsF6 P9olk8Gz6LDydX9SOZ4w/yPI8y8eU6z1AauwOPxw9F1lu82goIGwK3jIcvOv72koB5 Izq9By7L6PESEmmJ5nFc4ko9vH2CBMcJKPV95HTg= x-aol-sid: 3039ac1d338d52616bb37d53 Content-Length: 26445 The Email was from the aol domain, so I understand the IP of aol. My question is, looking at 66.199.226.81 , would it be safe to say that the Email originated from "Access Integrated Technologies"? Thanks for any help!

    Read the article

  • Rails 3 shows 404 error instead of index.html (nginx + unicorn)

    - by Miko
    I have an index.html in public/ that should be loading by default but instead I get a 404 error when I try to access http://example.com/ The page you were looking for doesn't exist. You may have mistyped the address or the page may have moved. This has something to do with nginx and unicorn which I am using to power Rails 3 When take unicorn out of the nginx configuration file, the problem goes away and index.html loads just fine. Here is my nginx configuration file: upstream unicorn { server unix:/tmp/.sock fail_timeout=0; } server { server_name example.com; root /www/example.com/current/public; index index.html; keepalive_timeout 5; location / { try_files $uri @unicorn; } location @unicorn { proxy_pass http://unicorn; proxy_set_header X-Forwarded-For $proxy_add_x_forwarded_for; proxy_set_header Host $host; proxy_redirect off; } } My config/routes.rb is pretty much empty: Advertise::Application.routes.draw do |map| resources :users end The index.html file is located in public/index.html and it loads fine if I request it directly: http://example.com/index.html To reiterate, when I remove all references to unicorn from the nginx conf, index.html loads without any problems, I have a hard time understanding why this occurs because nginx should be trying to load that file on its own by default. -- Here is the error stack from production.log: Started GET "/" for 68.107.80.21 at 2010-08-08 12:06:29 -0700 Processing by HomeController#index as HTML Completed in 1ms ActionView::MissingTemplate (Missing template home/index with {:handlers=>[:erb, :rjs, :builder, :rhtml, :rxml, :haml], :formats=>[:html], :locale=>[:en, :en]} in view paths "/www/example.com/releases/20100808170224/app/views", "/www/example.com/releases/20100808170224/vendor/plugins/paperclip/app/views", "/www/example.com/releases/20100808170224/vendor/plugins/haml/app/views"): /usr/local/rvm/gems/ruby-1.9.2-rc2/gems/actionpack-3.0.0.beta4/lib/action_view/paths.rb:14:in `find' /usr/local/rvm/gems/ruby-1.9.2-rc2/gems/actionpack-3.0.0.beta4/lib/action_view/lookup_context.rb:79:in `find' /usr/local/rvm/gems/ruby-1.9.2-rc2/gems/actionpack-3.0.0.beta4/lib/action_view/base.rb:186:in `find_template' /usr/local/rvm/gems/ruby-1.9.2-rc2/gems/actionpack-3.0.0.beta4/lib/action_view/render/rendering.rb:45:in `_determine_template' /usr/local/rvm/gems/ruby-1.9.2-rc2/gems/actionpack-3.0.0.beta4/lib/action_view/render/rendering.rb:23:in `render' /usr/local/rvm/gems/ruby-1.9.2-rc2/gems/haml-3.0.15/lib/haml/helpers/action_view_mods.rb:13:in `render_with_haml' etc... -- nginx error log for this virtualhost comes up empty: 2010/08/08 12:40:22 [info] 3118#0: *1 client 68.107.80.21 closed keepalive connection My guess is unicorn is intercepting the request to index.html before nginx gets to process it.

    Read the article

  • Exchange 2007 and migrating only some users under a shared domain name

    - by DomoDomo
    I'm in the process of moving two law firms to hosted Exchange 2007, a service that the consulting company I work for offers. Let's call these two firms Crane Law and Poole Law. These two firms were ONE firm just six months ago, but split. So they have three email domains: Old Firm: craneandpoole.com New Firm 1: cranelaw.com New Firm 2: poolelaw.com Both Firm 1 & Firm 2 use craneandpoole.com email addresses, as for the other two domains, only people who work at the respective firm use that firm's domain name, natch. Currently these two firms are still using the same pre-split internal Exchange 2007 server, where MX records for all three domains point. Here's the problem. I'm not moving both companies at the same time. I'm moving Crane Law two weeks before Poole Law. During this two weeks, both companies need to be able to: Continue to receive emails addressed to craneandpoole.com Send emails between firms, using cranelaw.com and poolelaw.com accounts I also have a third problem: I'd like to setup all three domains in my hosting infrastructure way ahead of time, to make my own life easier What would solve all my problems would be, if there is some way I can tell Exchange 2007, even though this domain exists locally forward on the message to the outside world using public MX record as a basis for where to send it (or if I could somehow create a route for it statically that would work too). If this doesn't work, to address points #1 when I migrate Crane Law, I will delete all references locally to cranelaw.com on their current Exchange server, and setup individual forwards for each of their craneandpool.com mailboxes to forward to our hosted exchange server. This will also take care of point #2, since the cranelaw.com won't be there locally, when poolelaw.com tries to send to cranelaw.com, public MX records will be used for mail routing decisions and go to my hosted exchange. The bummer of that though is, I won't be able to setup poolelaw.com ahead of time in hosted Exchange, will have to wait to do it day of :( Sorry for the long and confusing post. Just wondering if there is a better or simpler way to do what I want? Three tier forests and that kind of thing are out, this is just a two week window where they won't be in the same place.

    Read the article

  • Why would a PCI scan fail because of components that are not even installed?

    - by Brandon
    Recently a PCI scan was run against a web server and the result was a failure. Some of the issues could be fixed, however others simply make no sense to me. The machine was a clean install, there are only two things running, the .NET 3.5 website and the dotDefender web application firewall. However there are several errors similar to: Web server vulnerability Impact: /servlet/SessionServlet: JRun or Netware WebSphere default servlet found. All default code should be removed from servers. Risk Factor: Medium/ CVSS2 Base Score: 6.4 CVE: CVE-2000-0539 I'm not sure what this is, but I can't find anything on the server that looks anything like this. Web server vulnerability Impact: /some.php?=PHPE9568F35- D428-11d2-A769-00AA001ACF42: PHP reveals potentially sensitive information via certain HTTP requests that contain specific QUERY strings. Risk Factor: Medium/ CVSS2 Base Score: 5.0 PHP is not installed. Trying to add that query string to any page does nothing because the application ignores it. And doing that phpVersion check results in a 404. Similar to this, there are dozens of errors related to JSP and Oracle that are also not installed. Web server vulnerability Impact: /admin/database/wwForum.mdb: Web Wiz Forums pre 7.5 is vulnerable to Cross-Site Scripting attacks. Default login/pass is Administrator/letmein Risk Factor: Medium/ CVSS2 Base Score: 4.0 There are several errors like this, telling me that Web Wiz Forums, Alan Ward A-Cart 2.0, IlohaMail, etc. are all vulnerable. These are not installed or referenced anywhere I can find. There are even references to pages that simply don't exist, like OpenAutoClassifieds. Can anyone point me in the right direction as to why these errors are showing up or where I might look to find these components if they are in fact installed? Note: This website and server are for a subdomain of the main website. The main website runs on a server that is running Apache/PHP, but I don't have access to that server. The report says the subdomain was the site being scanned, but is it possible for it to have scanned the main site as well?

    Read the article

  • Coldfusion 8 Application Crashes Under Heavy Load

    - by KM01
    Hello, We have a CF8 app that runs for 20-25 minutes before crashing under heavy load ~ 1200 users. This load is generated by our load testing tool: 1200 users ramped up in 5 mins (approx behavior of our users), running for an hour. We have this app on Solaris 10, Apache 2, JRun 4 and Oracle 10g. Java version is 1.6. During the initial load tests, the thread dumps pointed to monitor deadlocks that pointed to sessions. "jrpp-173": waiting to lock monitor 0x019fdc60 (object 0x6b893530, a java.util.Hashtable), which is held by "scheduler-1" "scheduler-1": waiting to lock monitor 0x026c3ce0 (object 0x6abe2f20, a coldfusion.monitor.memory.SessionMemoryMonitor$TopMemoryUsedSessions), which is held by "jrpp-167" "jrpp-167": waiting to lock monitor 0x019fdc60 (object 0x6b893530, a java.util.Hashtable), which is held by "scheduler-1" We increased the number of sessions relative to the number of CPUs (48 simultaneous threads against 32 CPUs), and the deadlock went away. While varying the simultaneous threads helped a little bit in terms of response time, the CF server still tanked in 20-25 minutes during all of these tests. We ran more thread dumps, and saw a thread locking a monitor, for e.g.: "jrpp-475" prio=3 tid=0x02230800 nid=0x2c5 runnable [0x4397d000] java.lang.Thread.State: RUNNABLE at java.util.HashMap.getEntry(HashMap.java:347) at java.util.HashMap.containsKey(HashMap.java:335) at java.util.HashSet.contains(HashSet.java:184) at coldfusion.monitor.memory.MemoryTracker.onAddObject(MemoryTracker.java:124) at coldfusion.monitor.memory.MemoryTrackerProxy.onReplaceValue(MemoryTrackerProxy.java:598) at coldfusion.monitor.memory.MemoryTrackerProxy.onPut(MemoryTrackerProxy.java:510) at coldfusion.util.CaseInsensitiveMap.put(CaseInsensitiveMap.java:250) at coldfusion.util.FastHashtable.put(FastHashtable.java:43) - locked <0x6f7e1a78> (a coldfusion.runtime.Struct) at coldfusion.runtime.CfJspPage._arrayset(CfJspPage.java:1027) at coldfusion.runtime.CfJspPage._arraySetAt(CfJspPage.java:2117) at cfvalidation2ecfc1052964961$funcSETUSERAUDITDATA.runFunction(/app/docs/apply/cfcs/validation.cfc:377) As you see in the last line above there were several references CFMs and CFCs, and the lines have "cflock" tags, which were scoped to the "application." We (the dev team) then changed them to be scoped to a "name". After more load tests, there is no locking going on and there no deadlocks, but now the application tanks in 7-10 minutes. We've gotten system, network and DB reports from the respective admins, and they are not being taxed; even watched the server stats with server monitor, top, prstat, ran sar reports, etc. So we believe it is an issue with the CF server or maybe the JVM. I am running out of ideas as to what else we can try. Disclaimer: I am not a CF developer or Admin. I am just running the load test, analyzing the reports, threads etc, and sharing the results with the dev and admin teams, and trying the next change, and so on. So far no dice. Has anyone run into something similar? How did you go about diagnosing and troubleshooting? All thoughts and pointers welcome. Thank you for your time! KM

    Read the article

  • Bypassing "Found New Hardware Wizard" / Setting Windows to Install Drivers Automatically

    - by Synetech inc.
    Hi, My motherboard finally died after the better part of a decade, so I bought a used system. I put my old hard-drive and sound-card in the new system, and connected my old keyboard and mouse (the rest of the components—CPU, RAM, mobo, video card—are from the new system). I knew beforehand that it would be a challenge to get Windows to boot and install drivers for the new hardware (particularly since the foundational components are new), but I am completely unable to even attempt to get through the work of installing drivers for things like the video card because the keyboard and mouse won't work (they do work, in the BIOS screen, in DOS mode, in Windows 7, in XP's boot menu, etc., just not in Windows XP itself). Whenever I try to boot XP (in normal or safe mode), I get a bunch of balloons popping up for all the new hardware detected, and a New Hardware Found Wizard for Processor (obviously it has to install drivers for the lowest-level components on up). Unfortunately I cannot click Next since the keyboard and mouse won't work yet because the motherboard drivers (for the PS/2 or USB ports) are not yet installed. I even tried a serial mouse, but to no avail—again, it does work in DOS, 7, etc., but not XP because it doesn't have the serial port driver installed. I tried mounting the SOFTWARE and SYSTEM hives under Windows 7 in order to manually set the "unsigned drivers warning" to ignore (using both of the driver-signing policy settings that I found references to). That didn't work; I still get the wizard. They are not even fancy, proprietary, third-party, or unsigned drivers. They are drivers that come with Windows—as the drivers for CPU, RAM, IDE controller, etc. tend to be. And the keyboard and mouse drivers are the generic ones at that (but like I said, those are irrelevant since the drivers for the ports that they are connected to are not yet installed). Obviously at some point in time over the past several years, a setting got changed to make Windows always prompt me when it detects new hardware. (It was also configured to show the Shutdown Event Tracker on abnormal shutdowns, so I had to turn that off so that I could even see the desktop.) Oh, and I tried deleting all of the PNF files so that they get regenerated, but that too did not help. Does anyone know how I can reset Windows to at least try to automatically install drivers for new hardware before prompting me if it fails? Conversely, does anyone know how exactly one turns off automatic driver installation (and prompt with the wizard)? Thanks a lot.

    Read the article

  • How do large blobs affect SQL delete performance, and how can I mitigate the impact?

    - by Max Pollack
    I'm currently experiencing a strange issue that my understanding of SQL Server doesn't quite mesh with. We use SQL as our file storage for our internal storage service, and our database has about half a million rows in it. Most of the files (86%) are 1mb or under, but even on fresh copies of our database where we simply populate the table with data for the purposes of a test, it appears that rows with large amounts of data stored in a BLOB frequently cause timeouts when our SQL Server is under load. My understanding of how SQL Server deletes rows is that it's a garbage collection process, i.e. the row is marked as a ghost and the row is later deleted by the ghost cleanup process after the changes are copied to the transaction log. This suggests to me that regardless of the size of the data in the blob, row deletion should be close to instantaneous. However when deleting these rows we are definitely experiencing large numbers of timeouts and astoundingly low performance. In our test data set, its files over 30mb that cause this issue. This is an edge case, we don't frequently encounter these, and even though we're looking into SQL filestream as a solution to some of our problems, we're trying to narrow down where these issues are originating from. We ARE performing our deletes inside of a transaction. We're also performing updates to metadata such as file size stats, but these exist in a separate table away from the file data itself. Hierarchy data is stored in the table that contains the file information. Really, in the end it's not so much what we're doing around the deletes that matters, we just can't find any references to low delete performance on rows that contain a large amount of data in a BLOB. We are trying to determine if this is even an avenue worth exploring, or if it has to be one of our processes around the delete that's causing the issue. Are there any situations in which this could occur? Is it common for a database server to come to the point of complete timeouts when many of these deletes are occurring simultaneously? Is there a way to combat this issue if it exists? (cross-posted from StackOverflow )

    Read the article

  • What is the current state of Ubuntu's transition from init scripts to Upstart? [migrated]

    - by Adam Eberlin
    What is the current state of Ubuntu's transition from init.d scripts to upstart? I was curious, so I compared the contents of /etc/init.d/ to /etc/init/ on one of our development machines, which is running Ubuntu 12.04 LTS Server. # /etc/init.d/ # /etc/init/ acpid acpid.conf apache2 --------------------------- apparmor --------------------------- apport apport.conf atd atd.conf bind9 --------------------------- bootlogd --------------------------- cgroup-lite cgroup-lite.conf --------------------------- console.conf console-setup console-setup.conf --------------------------- container-detect.conf --------------------------- control-alt-delete.conf cron cron.conf dbus dbus.conf dmesg dmesg.conf dns-clean --------------------------- friendly-recovery --------------------------- --------------------------- failsafe.conf --------------------------- flush-early-job-log.conf --------------------------- friendly-recovery.conf grub-common --------------------------- halt --------------------------- hostname hostname.conf hwclock hwclock.conf hwclock-save hwclock-save.conf irqbalance irqbalance.conf killprocs --------------------------- lxc lxc.conf lxc-net lxc-net.conf module-init-tools module-init-tools.conf --------------------------- mountall.conf --------------------------- mountall-net.conf --------------------------- mountall-reboot.conf --------------------------- mountall-shell.conf --------------------------- mounted-debugfs.conf --------------------------- mounted-dev.conf --------------------------- mounted-proc.conf --------------------------- mounted-run.conf --------------------------- mounted-tmp.conf --------------------------- mounted-var.conf networking networking.conf network-interface network-interface.conf network-interface-container network-interface-container.conf network-interface-security network-interface-security.conf newrelic-sysmond --------------------------- ondemand --------------------------- plymouth plymouth.conf plymouth-log plymouth-log.conf plymouth-splash plymouth-splash.conf plymouth-stop plymouth-stop.conf plymouth-upstart-bridge plymouth-upstart-bridge.conf postgresql --------------------------- pppd-dns --------------------------- procps procps.conf rc rc.conf rc.local --------------------------- rcS rcS.conf --------------------------- rc-sysinit.conf reboot --------------------------- resolvconf resolvconf.conf rsync --------------------------- rsyslog rsyslog.conf screen-cleanup screen-cleanup.conf sendsigs --------------------------- setvtrgb setvtrgb.conf --------------------------- shutdown.conf single --------------------------- skeleton --------------------------- ssh ssh.conf stop-bootlogd --------------------------- stop-bootlogd-single --------------------------- sudo --------------------------- --------------------------- tty1.conf --------------------------- tty2.conf --------------------------- tty3.conf --------------------------- tty4.conf --------------------------- tty5.conf --------------------------- tty6.conf udev udev.conf udev-fallback-graphics udev-fallback-graphics.conf udev-finish udev-finish.conf udevmonitor udevmonitor.conf udevtrigger udevtrigger.conf ufw ufw.conf umountfs --------------------------- umountnfs.sh --------------------------- umountroot --------------------------- --------------------------- upstart-socket-bridge.conf --------------------------- upstart-udev-bridge.conf urandom --------------------------- --------------------------- ureadahead.conf --------------------------- ureadahead-other.conf --------------------------- wait-for-state.conf whoopsie whoopsie.conf To be honest, I'm not entirely sure if I'm interpreting the division of responsibilities properly, as I didn't expect to see any overlap (of what framework handles which services). So I was quite surprised to learn that there was a significant amount of overlap in service references, in addition to being unable to discern which of the two was intended to be the primary service framework. Why does there seem to be a fair amount of redundancy in individual service handling between init.d and upstart? Is something else at play here that I'm missing? What is preventing upstart from completely taking over for init.d? Is there some functionality that certain daemons require which upstart does not yet have, which are preventing some services from converting? Or is it something else entirely?

    Read the article

  • Unable to load type error in ASP.NET 4 application on Windows Server 2003 / IIS6 -- only happens after first worker process recycle

    - by Daniel Coffman
    I'm running an ASP.NET 4.0 web application on IIS6 (Windows Server 2003 x64). This app is one of many running on this server under the Default Web Site -- but is alone on it's own application pool because the other sites are all running ASP.NET 2.0 still. When I deploy my application, it works just fine until the application pool recycles or kills its worker process (by default 2 hours or 20 minutes with no activity). After this, I get the error: "Unable to load one or more of the requested types. Retrieve the LoaderExceptions property for more information. System.Reflection.ReflectionTypeLoadException: Unable to load one or more of the requested types. Retrieve the LoaderExceptions property for more information." Refreshing the page, recycling the application pool, and iisreset do nothing. But, I can bring the site back online again for a little while by simply redeploying it. The stack trace seems to start at an EntityDataSource -- see below: [ReflectionTypeLoadException: Unable to load one or more of the requested types. Retrieve the LoaderExceptions property for more information.] System.Reflection.RuntimeModule.GetTypes(RuntimeModule module) +0 System.Reflection.Assembly.GetTypes() +144 System.Data.Metadata.Edm.ObjectItemConventionAssemblyLoader.LoadTypesFromAssembly() +45 System.Data.Metadata.Edm.ObjectItemAssemblyLoader.Load() +34 System.Data.Metadata.Edm.AssemblyCache.LoadAssembly(Assembly assembly, Boolean loadReferencedAssemblies, ObjectItemLoadingSessionData loadingData) +130 System.Data.Metadata.Edm.AssemblyCache.LoadAssembly(Assembly assembly, Boolean loadReferencedAssemblies, KnownAssembliesSet knownAssemblies, EdmItemCollection edmItemCollection, Action`1 logLoadMessage, Object& loaderCookie, Dictionary`2& typesInLoading, List`1& errors) +248 System.Data.Metadata.Edm.ObjectItemCollection.LoadAssemblyFromCache(ObjectItemCollection objectItemCollection, Assembly assembly, Boolean loadReferencedAssemblies, EdmItemCollection edmItemCollection, Action`1 logLoadMessage) +580 System.Data.Metadata.Edm.ObjectItemCollection.ExplicitLoadFromAssembly(Assembly assembly, EdmItemCollection edmItemCollection, Action`1 logLoadMessage) +193 System.Data.Metadata.Edm.MetadataWorkspace.ExplicitLoadFromAssembly(Assembly assembly, ObjectItemCollection collection, Action`1 logLoadMessage) +140 System.Web.UI.WebControls.EntityDataSourceView.ConstructContext() +756 System.Web.UI.WebControls.EntityDataSourceView.ExecuteSelect(DataSourceSelectArguments arguments) +147 This is a bug filed for the same (or similar) problem: http://connect.microsoft.com/VisualStudio/feedback/details/541962/unable-to-load-one-or-more-of-the-requested-types-connected-with-entitydatasource Question: Has anyone seen this and have advice? I've tried copy-local on all the references... Works just fine on my dev machine. Works on the server until the application pool worker process recycles. I'm building in release mode, but experience the same result when I build for debug. I'm stumped.

    Read the article

  • Moving from VPS to Cloud

    - by GRIGORE-TURBODISEL
    ...and I have a few questions. I'm basically working on a MySQL+PHP based webapp. Since I don't have on-demand scaling with VPS, I'm planning to move from VPS to Cloud when I hit the 1000 subscribers barrier. I'm looking at Windows Azure but I'm ok with other suggestions. So here are my questions: Will it really cost me a kidney? Every subscriber needs to download around 4-5MB of static resources each day. Bandwidth is free on the VPS but here I see costs can easily get to $800.00/mo; this makes me very insecure about the whole thing, I mean VPS is just $2,000/yr. Do I need another VM or is PHP included in the Web Sites? I have basic sysadmin skills, I think I can handle setting up a PHP install, but will I have to do this? If yes, what other service do I need to setup manually? What about Memcached, MySQL, etc? What security protections does it include? For example I have some basic protection included, like directory traversals and executable files upload; I also have CloudFlare on my other websites for DDoS protection; will I need to do the same thing here too, can it even be installed, can I edit my DNS records, etc? How are e-mails, subdomains, add-on domains, parked domains, etc. handled? I haven't seen any references to e-mail boxes. On the VPS I simply add them from cPanel ([email protected] / whatever.mysite.com / ...); do I have a similar management interface here? Do I get SSH access? Or at least FTP, remote MySQL access and maybe some incremental back-ups or something? Can I see my quotas and advanced traffic info? I must mention that I really like the idea of the whole "cloud" concept, the added reliability and everything but I really need maybe a parallel to regular hosting or something so I know what to expect.

    Read the article

  • APC PHP cache size does not exceed 32MB, even though settings allow for more

    - by hardy101
    I am setting up APC (v 3.1.9) on a high-traffic WordPress installation on CentOS 6.0 64 bit. I have figured out many of the quirks with APC, but something is still not quite right. No matter what settings I change, APC never actually caches more than 32MB. I'm trying to bump it up to 256 MB. 32MB is a default amount for apc.shm_size, so I am wondering if it's stuck there somehow. I have run the following echo '2147483648' > /proc/sys/kernel/shmmax to increase my system's shared memory to 2G (half of my 4G box). Then ran ipcs -lm which returns ------ Shared Memory Limits -------- max number of segments = 4096 max seg size (kbytes) = 2097152 max total shared memory (kbytes) = 8388608 min seg size (bytes) = 1 Also made a change in /etc/sysctl.conf then ran sysctl -p to make the settings stick on the server. Rebooted, too, for good measure. In my APC settings, I have mmap enabled (which happens by default in recent versions of APC). php.ini looks like: apc.stat=0 apc.shm_size="256M" apc.max_file_size="10M" apc.mmap_file_mask="/tmp/apc.XXXXXX" apc.ttl="7200" I am aware that mmap mode will ignore references to apc.shm_segments, so I have left it out with default 1. phpinfo() indicates the following about APC: Version 3.1.9 APC Debugging Disabled MMAP Support Enabled MMAP File Mask /tmp/apc.bPS7rB Locking type pthread mutex Locks Serialization Support php Revision $Revision: 308812 $ Build Date Oct 11 2011 22:55:02 Directive Local Value apc.cache_by_default On apc.canonicalize O apc.coredump_unmap Off apc.enable_cli Off apc.enabled On On apc.file_md5 Off apc.file_update_protection 2 apc.filters no value apc.gc_ttl 3600 apc.include_once_override Off apc.lazy_classes Off apc.lazy_functions Off apc.max_file_size 10M apc.mmap_file_mask /tmp/apc.bPS7rB apc.num_files_hint 1000 apc.preload_path no value apc.report_autofilter Off apc.rfc1867 Off apc.rfc1867_freq 0 apc.rfc1867_name APC_UPLOAD_PROGRESS apc.rfc1867_prefix upload_ apc.rfc1867_ttl 3600 apc.serializer default apc.shm_segments 1 apc.shm_size 256M apc.slam_defense On apc.stat Off apc.stat_ctime Off apc.ttl 7200 apc.use_request_time On apc.user_entries_hint 4096 apc.user_ttl 0 apc.write_lock On apc.php reveals the following graph, no matter how long the server runs (cache size fluctuates and hovers at just under 32MB. See image http://i.stack.imgur.com/2bwMa.png You can see that the cache is trying to allocate 256MB, but the brown piece of the pie keeps getting recycled at 32MB. This is confirmed as refreshing the apc.php page shows cached file counts that move up and down (implying that the cache is not holding onto all of its files). Does anyone have an idea of how to get APC to use more than 32 MB for its cache size?? **Note that the identical behavior occurs for eaccelerator, xcache, and APC. I read here: http://www.litespeedtech.com/support/forum/archive/index.php/t-5072.html that suEXEC could cause this problem.

    Read the article

< Previous Page | 142 143 144 145 146 147 148 149 150 151 152 153  | Next Page >