I'm not sure exactly how to phrase the question. But, I'm researching how to PXE boot a server without having a DHCP/PXE server in each vlan.
Scenario: Datacenter with dozens of servers. 1 VLAN per server. Cisco switches and routers. Each server has a serial console available for remote management (OS and BIOS are configured for serial console). If an admin wants to re-install OS, they should be able to reboot the server and tell the BIOS to initiate a PXE boot request. A central install server is available to provide the DHCP and PXE boot images.
Has anyone tried this? I have been reading about the 'ip helper-address' for Cisco to relay DHCP requests. Interested in hearing about real-world setups. Or is there a better way to accomplish remote OS installs?
We are co-located at a datacenter and host web sites, and corporate email systems, as well as host dedicated servers for customers.
We currently have two /25 internet facing subnets from our provider. We have a Watchguard X5500e 8 port gigabit firewall that supports routing as well as VLANs. We also currently own QTY4 2848 HP Gigabit switches.
We currently have each switch connected in a loop with 2 gigabit ports trunked using static LCAP. The switches are connected as follows: A > B B > C C > D D > A
Rapid STP is turned on. One thing is - is this the ideal trunking scheme?
The more important question is this. We would like to separate ips from each other using VLANing.
IE: we might have a client with 5 different IPs in one or more subnets and we would like to group them together.
We ideally do not want to break up the subnets into smaller ones as it makes it hard to reconfigure and it wastes ip addresses, as we do not have that many.
Is there a VPS provider that will sell me a VM , and put it up somewhere, and can make me another VM in the future, on the same VLAN as the original VM ?
For example, pretend VM #1 has a NIC at 10.0.0.100
in the future, I want another VM with a nic at 10.0.0.101
I was looking at Go-Grid , but I'm not sure how their pricing works.
I need a basic L3 switch for maybe 25 mbps that will do hopefully up to 50 VLANs and which will not require me to hire someone to configure it.
As much as I like Cisco, that rules them out.
The reason I'd like a Layer 3 switch is so that I can run my backups and inter-server transfers without adding to my bandwidth bill. Also, VLANS are a critical requirement as i have a lot of customers with root on their managed servers.
So i am looking at HP [gasp] switches. How "easy" is the web-based configuration widget? [I'm an advanced unix admin but networking is a mystery to me.]
This is a starter switch and once i have a full cab of servers I'll be able to spend $7K on a pair of 3560s and hire someone to configure them for me ... but until then what can i get to meet my requirements?
I'm trying to implement VLANs on my network and can't get connectivity to host servers. Here's how the network is configured. Pardon the bad ascii diagram.
In this example my upstream is providing two subnets:
111.111.111.16/28 (I'm using an IP from this subnet to manage the 3550)
222.222.222.16/29
I am attempting subdivide the /29 into two /30's in order to place a server into it's own /30 subnet & VLAN ............
I have two servers both in a same vlan. Both may access Internet and be acceessed from Internet I setup db server and web server internal IP each as follows:
I used ifconfig to check both status, both of them are up. both of them may ping google, but when I try to ping their each other through internal IP, nothing returns.
I used command tracert to follow, found all packages were sent to Internet rather than an internal IP.
My host tells me to do it by making NAT, I have no idea on it. Anyone may help me out on how to do with NAT?
I'm looking for a solution that I can place a firewall between 2 vlans on a BigIron router with L3 enabled.
For this moment there is one big vlan2 with a ip-route 0.0.0.0 0.0.0.0 123.123.123.123 and a router-interface ve2 with the IP of the router, the address I use as gateway on the machines behind it.
The WAN port has the IP address to communicate with to the GW of the carrier-router (123.123.123.122)
Because I want to let the BigIron the routing I was thinking of 2 vlans, one for the lan-vlan and one for the wan-vlan, but this will be a problem because I only have one IP-block what I can use.
So the sitiuation must be as follow on the BigIron:
WAN => vlan2 => firewall => vlan3(lan)
Because of the fact that the firewall will be transparent, this should be no problem to place it between the vlans. The actual problem is how to manage this. In simple words, I should be able to replace the firewall with a cross-cable and it should still work.
Cisco for an example has a SVI solution for this, but I can't find such thing for a Foundry router.
Having a slight problem working with one of our Extreme Summit 48 (ugh) switches - I've figured out most of the basics, but I can't seem to find any way to add a secondary IP address to a VLAN! This, I would have thought, would be a pretty basic feature to have. Typing "config vlan [vlanname] ipaddress 1.2.3.4/24" works for setting the primary IP, but I can't figure out how to add any more - and doing the command again just overwrites the first one.
So... does anyone have any tricks up their sleeve, or is this something that Extreme neglected to add to this model switch?
We offer colocation & dedicated servers as well as shared & reseller hosting services.
Our colocation customers and dedicated server customers are definitely on their own VLANs for obvious reasons.
Up until now, we have been using separate VLANS and ip allocations for each of the servers in our shared & reseller server fleet. I'm starting to question this policy for many reasons:
1) We directly manage all of the servers and it is very rare that any servers are compromised to the point where they can steal an IP address.
2) We are wasting IP addresses - network, broadcast and gateway addresses are required for each vlan. Additionally, if a server needs 1 more IP address, we need to add a whole new block.
If all of the servers are under our direct management, does it make sense for us to use any vlans at all? It seems that it only serves to complicate things, waste ips and add management overhead.
I've read that all ethernet switches in a MST Region need the same Name, Revision number, and list of member vlans for each Instance. So what happens when you need to change the range of VLANs in a MSTI ? Let's say that you need to add a range of vlans to an instance that spans 20 switches? How would you do that?
Can you make a recommendation for a switch-based L3 router which can
- hold a moderate number of routes (interface routes, a few hundred statics + default) - OSPF and BGP - MST - 1024 layer-3 dot1q subinterfaces (or maybe VLAN interfaces) with + traffic policing in and out per subinterface/vlan + VRRP/HSRP/NSRP - IPv4 & IPv6 native - 2x GigE ports - Not tip-over under 1gbps DDoS towards a VLAN interface.
I've been using 3560Gs, but they seem to lack the output traffic policing. I'd prefer to have subinterfaces which don't run spanning-tree, versus Vlan Interfaces to a trunk interface which runs spanning-tree. These switches sit at the L3 boundary between two L2 networks.
Cost is a big factor; but I also must carry vendor licenses & support contract, if the vendor asserts that not doing so is illegal in US.
I've supposedly set APF firewall to start at boot time, by doing something like:
chkconfig --levels 2345 apf on
However i have my reservations to weather it is actually starting, its set to block port 80, after boot if i try and access it, the connection will get refused straight away, however if i go and manually start APF then try and access again, it will take a while, like its ignoring the connection attempt (which is good).
I'm trying to put together a new 1U rackmount server that I'm intending to colocate (my first server of my own). Here are the details:
11-152-087 CHASSIS SUPERMICRO|CSE-512L-260B R 13-182-120 SERVER_MB SUPERMICRO|MBD-PDSML-LN2+ 22-144-417 HD 250G|WD 7K 8M SATA2 WD2500JS 19-115-029 CPU INTEL|C2D E6750 2.66G 65N 4M R 20-134-337 MEM 1Gx2|KST KVR667D2E5K2/2G R
The stock CPU fan is obviously too big to sit in a 1U rackmount chassis, but I've got the top off and was just trying to boot up with the OEM fan to make certain everything worked (forgot to order a low-profile fan and heatsink--they're on the way).
Well, I put everything together, and it won't boot. Upon power up, there is nothing sent to the VGA port, and after a second or two, there is a hi-low alarm sound that does not stop that seems to imply the system is overheating (the MB and chassis manuals don't mention this alarm anywhere, but the read "temperature" light on the front is red). The CPU fan runs as well as the case fan, and even if I hold a huge high-velocity home fan over it, it still won't send any signal to the VGA monitor or POST.
I've checked that the power connections are good, the fans are seated properly, the memory contact is good, etc., and all I get is this hi-low chime every time I try to boot up. I think it's either a bad MB or CPU, since I can't imagine that the CPU is overheating with the heat sink and CPU fan.
Perhaps I've screwed something up, but the next issue I have is trying to figure out what could be causing the problem so I know what to return or replace. Is it the CPU? MB? RAM? Power unit? How can I test any of these without buying another set and possibly frying those?
I figured maybe some of you might have some ideas. I've built several computers over the years, but this one has me completely stymied.
This is probably the wrong location for this thread but I am unsure which section it is best located at.
I have 2 hard drives - 1 has XP pro installed (master drive) on it and the other is blank and set to slave. On the blank one i intedn to install Red Hat linux.
How is the best way to create a dual boot with these machines if each hard drive has one O/S dedicated to it.
We cannot figure out why our dedicated server will not boot to the correct kernel. I've removed all other options from grub.conf but it's still booting to the default CentOS setup.
grub.conf:
Code: # grub.conf generated by anaconda # # Note that you do not have to rerun grub after making changes to this file # NOTICE: You do not have a /boot partition. This means that # all kernel and initrd paths are relative to /, eg. # root (hd0,0) # kernel /boot/vmlinuz-version ro root=/dev/mapper/ddf1_4c53492020202020808627c300000000378494a900000a28p1 # initrd /boot/initrd-version.img #boot=/dev/mapper/ddf1_4c53492020202020808627c300000000378494a900000a28 default=0 timeout=5 splashimage=(hd0,0)/boot/grub/splash.xpm.gz hiddenmenu title CentOS (2.6.18-028stab062.3) root (hd0,0) kernel /boot/vmlinux-2.6.18-028stab062.3 ro root=LABEL=/ initrd /boot/initrd-2.6.18-028stab062.3.img
I've got Centos 4 and I'm wondering what's the best way to cleanup my /boot partition?
Tried to do a yum update tonight and it included kernel, amongst other updates that belonged there so it stopped. I've googled around for commands to run and whatnot, but no go... or I just can't find it... if I had to clean it up I have an idea already about what to do, but I want to ask for advice first to see if there's an easier way.
on a RHE 3 system I installed a new Kernel. I did update lilo.conf and grub.conf but they are still booting an old kernel: 2.4.21-27.ELsmp. Please have a look at my files below and if you have any idea why please let me know.
Code: default=0 timeout=10 splashimage=(hd0,0)/grub/splash.xpm.gz title Red Hat Enterprise Linux ES (2.4.21-53.ELsmp) root (hd0,0) kernel /vmlinuz-2.4.21-53.ELsmp ro root=/dev/hda3 initrd /initrd-2.4.21-53.ELsmp.img title Red Hat Enterprise Linux ES (2.4.21-47.ELsmp) root (hd0,0) kernel /vmlinuz-2.4.21-47.ELsmp ro root=/dev/hda3 initrd /initrd-2.4.21-47.ELsmp.img title Red Hat Enterprise Linux ES (2.4.21-47.EL) root (hd0,0) kernel /vmlinuz-2.4.21-47.EL ro root=/dev/hda3 initrd /initrd-2.4.21-47.EL.img title Red Hat Enterprise Linux ES (2.4.21-27.ELsmp) root (hd0,0) kernel /vmlinuz-2.4.21-27.ELsmp ro root=LABEL=/ initrd /initrd-2.4.21-27.ELsmp.img title Red Hat Enterprise Linux ES-up (2.4.21-27.EL) root (hd0,0) kernel /vmlinuz-2.4.21-27.EL ro root=LABEL=/ initrd /initrd-2.4.21-27.EL.img
am running CentOs with whm/cpanel and am whenever i reboot my server all sites that required mysql show erros ( can't connect ) it resolves by manually in the web host manager restarting mysql.
I have an old server with only IDE and SCSI connectors on the mobo.
However, I have only SATA drives and no IDE available. I consider to buy an entry level PCI SATA adapter card. What do you think ? Can I install and boot Linux directly from it?
