ianc1215's questions

I am working on a mock up of a PXE server for the place I work. I am trying to put the image files and configuration file in places that keep things a little more organized. Even though my dhcpd.confspecifies the configuration options needed to tell PXELinux where to look for config files it still does not find said files.

Here is what my /etc/dhcp/dhcpd.conf looks like, this is where I imagine the issue is.

option space PXE;
option PXE.mtftp-ip             code 1 = ip-address;
option PXE.mtftp-cport          code 2 = unsigned integer 16;
option PXE.mtftp-sport          code 3 = unsigned integer 16;
option PXE.mtftp-tmout          code 4 = unsigned integer 8;
option PXE.mtftp-delay          code 5 = unsigned integer 8;
option arch                     code 93 = unsigned integer 16;

option space pxelinux;
option pxelinux.magic                   code 208 = string;
option pxelinux.configfile              code 209 = text;
option pxelinux.pathprefix              code 210 = text;
option pxelinux.reboottime              code 211 = unsigned integer 32;

ddns-update-style none;

option domain-name "lab.solignis.com";
option domain-name-servers ns01.lab.solignis.com;

default-lease-time 600;
max-lease-time 7200;

authoritative;
log-facility local7;

subnet 192.168.147.0 netmask 255.255.255.0 {
        range 192.168.147.100 192.168.147.254;
        option subnet-mask 255.255.255.0;
        option routers 192.168.147.2;
        option broadcast-address 192.168.147.255;
        default-lease-time 600;
        max-lease-time 7200;

        class "pxeclients" {

                match if substring (option vendor-class-identifier, 0, 9) = "PXEClient";
                next-server 192.168.147.10;

                if option arch = 00:00 {
                        option pxelinux.configfile "common.cfg";
                        option pxelinux.pathprefix "/configs/";
                        filename "/pxelinux/images/BIOS/pxelinux.0";
                } elsif option arch = 00:06 {
                        #option pathprefix "/configs/";
                        #option configfile "/EFIx64/default";
                        filename "/images/EFIx64/syslinux.efi";
                } elsif option arch = 00:07 {
                        #option pathprefix "/configs/";
                        #option configfile "/EFIx64/default";
                        filename "/images/EFIx64/syslinux.efi";
                } elsif option arch = 00:09 {
                        #option pathprefix "/configs/";
                        #option configfile "/EFIx64/default";
                        filename "/images/EFIia32/syslinux.efi";
                }
        }

}

Parts of the config were compiled from various parts of documentation, the parts that are commented out are old pieces that I have not fixed yet disregard those. For now I just want to get standard BIOS working (arch = 00:00).

Here is what my file structure looks like for my tftpboot directory.

server01@server01:/var/lib/tftpboot$ tree
.
└── pxelinux
    ├── configs
    │   └── common.cfg
    └── images
        ├── BIOS
        │   ├── ldlinux.c32
        │   └── pxelinux.0
        ├── EFIia32
        │   └── syslinux.efi
        └── EFIx64
            └── syslinux.efi

6 directories, 5 files

I was pretty sure I have it setup right for pxelinux but it does not see the config file so something is not configured right. Can anyone provide an idea of what I misconfigured? I have been banging my head on this all day.

This might be a little odd but basically I have a domain that exists on the outside let's call it "deploy.myitcompany.com". This domain is used for deploying iPads for a company using Apple MDM. The issues I running into is inside the building they are using an entirely different domain. Let's call that one "abccompany.com".

I am looking for a way to setup a DNS forward lookup zone (in Windows DNS) and add ONLY the record I care about which would be "deploy.myitcompany.com" BUT if I don't know the entry I want to forward the request to my configured forwarders.

Does this make sense? The issue I run into is if I add "myitcompany.com" to the local DNS server then certain records like "www" and "mail" no longer resolve unless I add the records to the local DNS which I would like to avoid.

Any insight into this?

My original idea was to fix it by changing the entire DNS server to be more simple to work with but this is being met with resistance so I need a workaround.

Thanks!

UPDATE

Just to clarify a bit on what I meant.

I work as a contractor for a small IT company. The company was working with a client to deploy iPads on their network. The previous guy had setup the Mac server to point to deploy.myitcompany.com instead of the client's domain. The issue I am running into is if I put a domain in their DNS server that includes "myitcompany.com" and add an A record it works as it should. But now if I try to go to "www.myitcompany.com" inside the network it fails since the local DNS does not know where "www" is supposed to go since there is no record. I am able to override this issue by adding a "www" record but that would mean I have to add all of the records for our domain so they work from with in the clients network.

Since the client really owns the equipment what I meant by fixing the entire DNS server was to change everything to what "should" be for the client and in turn not having to add our records to their server. But as I said I am meeting some resistance on this.

I was looking for a way to do a kind of split DNS where the server holds some other records for the domain and anything it does not know how to fulfill I wanted it to pass to the configured forwarders I setup.

Does that make sense?

I have an HP DL360 G7 server I plan to install KVM and ZFS on. The purpose is going to be a lab in a box. I have the 8 drive bays in the front loaded with 4 drives and an SSD (for the ZFS ZIL cache). My goal was to keep the disk array away from the actual OS disk. What I am wonder is weather or not it is safe to boot a Linux server installation from a USB drive for "production" use. The server has an embedded USB / SD card reader on the motherboard for VMware and other embedded solutions. This raises a question to me because once VMware is loaded it stay in memory. On the other hand a Linux install does not (atleast not 100%). I am concerned if I load the OS on a USB drive (or SD card) I will burn the SD card out.

Can anyone please give me some insight on this? I am wondering what my options are. The way I see it my options currently are make Linux boot from the ZFS array or use a USB drive. The first option would be okay if I could make grub play nice with ZFS root booting. Which to my knowledge is broken or at least needs some nasty hack to pull it off.

Like I said any help you could provide would be great.

I bought an HP DL360 G7 1 U rackmount server. I am looking to put it in a colocation facility. The issue is almost every facility I have seen either limits the amount of power I can access or charge extra for more power over the base amount provided.

Speaking strictly from the server's power supply ratings I believe it said they draw 7 amps a piece and there are 2 of them. But I am under the impression that 7 amps is when the power supplies are running at full tilt.

How can determine what the load will be on the power supplies on average?

Would the server run happily of 2.5amps per power supply? That's what the colo will provide me.

I just recently bought a new server an HP DL380 G6. I replaced the stock smart array p410 controller with an LSI 9211-8i.

My plan is use ZFS as the underlying storage for XEN which will run on the same baremetal.

I have been told that you can use SATA disks with the smart array controllers but because consumer drives lack TLER, CCTL and ERC its not recommended. Is this the case?

I was wondering if using the LSI controller in JBOD (RAID passthrough mode) does the kind of disks I use really have as much of an impact as they would with the smart array controller?

I am aware that trying to use a RAID system not backed by a write cache for virtualization is not good for performance. But I was conisdering adding an SSD for ZFS. Would that make any difference?

I reason I am so obsessed with using ZFS is for dedup and compression. I don't think the smart array controller can do any of those features.

I am currently running VMware vSphere 5 on a server I use for various projects. Mainly I use it for honing my skills with administration and running a few game servers.

Recently I was considering a move to Xen (xen.org) or possibly Citrix XenServer (Free version).

When I was looking at the various features not included in XenServer I noticed that one feature not included was memory overcommit. To my limited knowledge this was / is on VMware vSphere. The problem is since I was using VMware in a small environment (1 server) I am not even sure if I was utilizing memory overcommit.

So I am wondering partially what it is (memory overcommit) and how do I know if I was utilizing it or not?

I am trying to use a Windows 2003 R2 server as a secondary dns server for my BIND 9 master. The problem is I use keys to control transfers and I dont know how to add the key to Windows so its able to copy the zone files from the master.

I saw a TechNet article about enabling DNSSEC which I also need.

Any idea how to make this work?

I am running my pfSense gateway in a ESXi server. right now the server gets its time from the NTP server specified in the general settings section of the WebGUI.

I want to disable this so pfSense will only get its time from the local system which is controlled by VMware which is being sync to a local NTP server.

If I look at the FreeBSD manual it talks about changing the value of ntpdate_enable = YES to ntpdate_enable = NO in /etc/rc.conf.

Does this apply to pfSense also?

Oh yeah, here is what I was looking at. FreeBSD Handbook

I had read on a few forums about pfSense that said it was dangerous to virtualize pfSense. The reason that was stated was an attacker could use pfsense as a spring board for an attack on the hypervisor and then use that to gain access to the other virtual machines and eventually take everything offline.

It sounds crazy to me but is there shred of reality in that idea? Is running a router in a virtual server a bad idea?

I am trying to restore my OpenLDAP server from an ldif file I made before it went horribly wrong.

All of the websites I found talk about using slapadd but when I run slapadd -v -l ~ns01/openldap_config.ldif it returns the error Could not stat config file "/etc/ldap/slapd.conf" : No such file or directory (2) slapadd: bad configuration file!

When I look in the directory indeed the file does not exist, but this is because Ubuntu 11.10 is running in RTC mode, my LDAP server uses the /etc/slapd.d/cn=config for the configuration. So? What am I missing? Is slapadd the wrong tool to use?

As I said above I running on Ubuntu 11.10 server edition 64-bit.

I am trying to write an init script for the SCST iSCSI target software for Debian / Ubuntu. The current one in the download does not work and seems to be outfitted to work on SuSE and others.

This what I have so far, but I am basing this off of my limited knowledge of init scripts and a lot of google searches.

I am just wondering if I am over thinking this, do I need to unload the modules everytime I stop the script. The modules to no load on startup, I know I need to start the iscsi daemon and load the config.

Any help would be appreciated.

#!/bin/sh

DAEMON=/usr/local/sbin/iscsi-scstd
PIDFILE=/var/run/iscsi-scstd.pid
CONFIG=/etc/scst/scst.conf

case "$1" in

start)
    echo "Starting SCST iSCSI target service: "
    modprobe scst
    modprobe scst_vdisk
    modprobe iscsi_scst
    start-stop-daemon --start --quiet --exec $DAEMON
    scstadmin -config $CONFIG
    $RETVAL=$?
    if [ $RETVAL == "0" ]; then
        echo -n "Success"
    fi
    ;;

stop)
    echo "Stopping SCST iSCSI target service: "
    modprobe -r scst
    modprobe -r scst_vdisk
    modprobe -r iscsi_scst
    start-stop-daemon --stop --quiet --exec $DAEMON --pidfile $PIDFILE
    $RETVAL=$?
    if [ $RETVAL == "3" ]; then
        echo -n "Success"
    fi
    if [ $RETVAL == "1" ]; then
        rm -f $PIDFILE
        echo -n "Success"
    fi
    ;;

restart)
    echo "Stopping SCST iSCSI target service: "
    modprobe -r scst
    modprobe -r scst_vdisk
    modprobe -r iscsi_scst
    start-stop-daemon --stop --quiet --exec $DAEMON --pidfile $PIDFILE
    sleep 2
    echo "Stopping SCST iSCSI target service: "
    modprobe scst
    modprobe scst_vdisk
    modprobe iscsi_scst
    start-stop-daemon --start --quiet --exec $DAEMON
    scstadmin -config $CONFIG
    ;;

status)
    $RETVAL=$?
    if [ $RETVAL == "0" ]; then
        echo -n "SCST iSCSI target service is running"
    fi
    if [ $RETVAL == "3" ]; then
        echo -n "SCST iSCSI target service is not running"
    fi
    if [ $RETVAL == "4" ]; then
        echo -n "SCST iSCSI target service is unknown"
    fi  


*)
    echo "Usage: /etc/init.d/scst {start|stop|restart}"
    exit 1

esac

exit 0

I actually ended up finding exactly what I was looking for in an old Ubuntu PPA for my iSCSI target. So this question is pretty much moot.

I just got done installing VMware vSphere 5 Free edition on my server. I noticed when I went to install the client that there was no option to install the host update utility.

How do you update the server now?

For my company we use vCenter, but what about the stnadalone VI client?

I need to upload a package to the server from HP, its some kind of driver, I can't find the utility.

I am having a bit of split in thought.

I am looking at getting a NAS for my company, we need it for our VMware server.

Originally the idea was to buy a NAS (a Netgear ReadyNAS)

But then I got some recommendations from the community (ServerFault) to build a NAS from a server and use something like NextentaStor, Openfiler or FreeNAS.

I have eliminated Openfiler for several reasons and I am now left between Nexentastor and FreeNAS.

The problem I am having is whether FreeNAS can be counted on for a production environment.

Is FreeNAS good enough to be an iSCSI back end to VMware? The last time I used FreeNAS it was version 0.7.

We would be using Nexentastor community not the paid version.

Any insight would be great.

Does anyone know if I can install normal 2.5" SATA drives inside an HP DL360 G5 server? Right now it has 6 72GB SAS drives. I am looking for raw storage for my files SAS performance in not required.

I am going to be implmenting a VMware vSphere 4 setup in our company. At first I had the plans of using a Netgear ReadyNAS business applicance for VM storage via iSCSI. But as I was looking around for parts for my VMware server (license 2/3) I was thinking of a different approach.

I was thinking what if I just loaded 6 WD Scorpio Black 500GB SATA drives into the server and ran a DAS setup rather than investing in a NAS.

The reason I was thinking we need more space than anything. We are most likely never going go beyond 1 VMware server because we are so small (my server does not count) We only will be running 4 VMs max. Also we do not have licenses for the VMware HA components.

What would be downsides to have the storage with the VM server in small setup like this?

Any advantages?

Backups would be handled by the vCenter server which is a separate server. So getting to the storage in either regard is not a problem.

Is it possible to setup PFsense to do IPv6 to IPv4?

I am setting up an IPv6 LAN for "testing" and also to mess around with and I can't get an IPv6 address from my ISP so I need a way to change IPv6 to IPv4.

Update

Ok I upgraded to the PFsense 2.0 RC1 which has IPv6 support.

Under the "Advanced Settings" there are options for IPv4 NAT encapsulation of IPv6 packets.

There is a checkbox to enable the option and also a entry box that says "IP Address"

What do I enter there?

I was reading some articles on VMware ESXi about storage. One thing mentioned was the advantages of being able to use vMotion, DRS and etc. to achieve High Availability. The one thing that I kept seeing was the word "Shared Storage".

What does this mean?

Right now I am running my ESXi server with an iSCSI backend (single linux server).

Is that considered shared storage, even though only one server can connect to a single target.

Considering NFS allows many connections to the same data does it have advantages over iSCSI?

Can someone gives me an idea?

I am thinking about converting my iSCSI server to NFS.

Also does ESXi (4.0+) support NFSv4?

I was playing with my new syslog server and had my m0n0wall firewall logs forwarded as a test, I noticed a bunch of recent firewall log entries that say that it blocked other WAN IPs from my ISP (I checked) from connecting to me on TCP port 445. Why would a random computer be trying to connect to me on a port apparently used for Windows SMB shares? Just internet garbage? A port scan?

Here is what I am seeing:

Mar 15 23:38:41 gateway/gateway ipmon[121]: 23:38:40.614422 fxp0 @0:19 b 98.82.198.238,60653 -> 98.103.xxx.xxx,445 PR tcp len 20 48 -S IN broadcast
Mar 15 23:38:42 gateway/gateway ipmon[121]: 23:38:41.665571 fxp0 @0:19 b 98.82.198.238,60665 -> 98.103.xxx.xxx,445 PR tcp len 20 48 -S IN
Mar 15 23:38:43 gateway/gateway ipmon[121]: 23:38:43.165622 fxp0 @0:19 b 98.82.198.238,60670 -> 98.103.xxx.xxx,445 PR tcp len 20 48 -S IN broadcast
Mar 15 23:38:44 gateway/gateway ipmon[121]: 23:38:43.614524 fxp0 @0:19 b 98.82.198.238,60653 -> 98.103.xxx.xxx,445 PR tcp len 20 48 -S IN broadcast
Mar 15 23:38:44 gateway/gateway ipmon[121]: 23:38:43.808856 fxp0 @0:19 b 98.82.198.238,60665 -> 98.103.xxx.xxx,445 PR tcp len 20 48 -S IN
Mar 15 23:38:44 gateway/gateway ipmon[121]: 23:38:43.836313 fxp0 @0:19 b 98.82.198.238,60670 -> 98.103.xxx,xxx,445 PR tcp len 20 48 -S IN broadcast
Mar 15 23:38:48 gateway/gateway ipmon[121]: 23:38:48.305633 fxp0 @0:19 b 98.103.22.25 -> 98.103.xxx.xxx PR icmp len 20 92 icmp echo/0 IN broadcast
Mar 15 23:38:48 gateway/gateway ipmon[121]: 23:38:48.490778 fxp0 @0:19 b 98.103.22.25 -> 98.103.xxx.xxx PR icmp len 20 92 icmp echo/0 IN
Mar 15 23:38:48 gateway/gateway ipmon[121]: 23:38:48.550230 fxp0 @0:19 b 98.103.22.25 -> 98.103.xxx.xxx PR icmp len 20 92 icmp echo/0 IN broadcast
Mar 15 23:43:33 gateway/gateway ipmon[121]: 23:43:33.185836 fxp0 @0:19 b 98.86.34.225,64060 -> 98.103.xxx.xxx,445 PR tcp len 20 48 -S IN broadcast
Mar 15 23:43:34 gateway/gateway ipmon[121]: 23:43:33.405137 fxp0 @0:19 b 98.86.34.225,64081 -> 98.103.xxx.xxx,445 PR tcp len 20 48 -S IN
Mar 15 23:43:34 gateway/gateway ipmon[121]: 23:43:33.454384 fxp0 @0:19 b 98.86.34.225,64089 -> 98.103.xxx.xxx,445 PR tcp len 20 48 -S IN broadcast

I blacked out part of my IP address for my own safety.

I would like to setup a syslog server to forward all log file from all of my VMs and servers. I really don't much about what is out there. So I turn to the community, Something on Linux is fine, what I want more is alert ability like emails telling me something is not right. If there was something to sort the logs by source that would be cool.

Where would I want run the syslog server from? My admin WS or a server/VM?

Any input would be wonderful.

Thanks in advance.

I am trying to make a script that will ping an IP address and tell you the name of the server it is pinging.

It works....mostly

I cannot seem to figure out how to make Linux (Ubuntu more specifically) to display the dns domain suffix that is set inside of the "resolv.conf" file.

if I type "hostname" at a command prompt all I get is "ubuntu01"

I tried typing domainname and all I get is the output "(none)" which is confusing me considering I can type the hsotname of any computer on my network and it will append the DNS suffix of "solignis.local" to it just like it is supposed to.

Any ideas?