4,690 reputation
1430
bio website jeddaniels.com
location California
age 36
visits member for 4 years, 9 months
seen 1 hour ago

"All right, guys, uh, listen. This is a blues riff in "B", watch me for the changes, and try and keep up?"


Jul
22
comment Allocated Memory difference in xen guests with Linux 2.6.18 and 2.6.32
On question 1, in the section below where you have the text "despite showing", you then go on to show that dmesg is reporting "500984k/524288k". That means "500984k is available out of 524288k total". The free command also shows 524288k total mem, as per your output.
Jul
22
comment Allocated Memory difference in xen guests with Linux 2.6.18 and 2.6.32
Centos 6.3 is also RedHat based, so will be different from Debian based (Ubuntu). Additionally, Centos 6.3 probably has different defaults compiled into the kernel than either 5.7 or Ubuntu, so it makes sense that it is different.
Jul
22
comment Why does OS X spend one more packet to establish and close a TCP connection?
tcpdump is perfectly valid for looking at this. The information is not easier to access in wireshark if you have to 1) install a GUI, 2) install wireshark.
Jul
22
comment Why VM cannot get the public IP I have from provider?
How about an ifconfig -a on the VM?
Jul
20
comment Reviewing firewall rules
RedSeal is more than just firewall rules, they do all kinds of stuff to help visualize your network security architecture, including determining if your current access control stance (implemented via firewalls, ACLs, etc.) is complaint with your desired company "policy" (or other standards, such as PCI-DSS, et al). As far as firewall rules go, they can analyze for a bunch of Best Practice checks, in addition to some firewall specific stuff such as unused rules, missing logging, etc. for a variety of different vendors.
Jun
11
comment How to restore Ubuntu server on a VMWare image after disk failure?
You probably need to mount specific partitions: "/dev/sdb3" or similar. Try "sudo fdisk -l /dev/sdb" to see what might be available.
Jun
10
comment How to restore Ubuntu server on a VMWare image after disk failure?
If you really have to "recover" it, i.e. make the actual same image bootable again, then yea, probably run the rescue process. Otherwise, just copy off your configs and data and you are all set (I've found that this tends to get much more reliable and consistent results, and I don't end up spending hours troubleshooting weird boot parameters and grub issues that I don't really understand.)
Jun
10
comment How does server handle multiple clients on the same port? Multiple sockets or multiple ports?
I second the recommendation for TCP/IP Illustrated.
May
28
comment How to connect the virtual networks of vmware guests running on different hosts?
Do the VMs all need to communicate with each other? Or just the host(s)?
May
28
comment How to connect the virtual networks of vmware guests running on different hosts?
Do the VMs all need to be on the same broadcast domain (i.e. subnet)?
May
26
comment GRE tunneling in vmware vsphere guest
The answer to this question is "Yes". But a one word answer is too short for a real answer, so I'm suggesting that you elaborate on your question with details on why you think it might not work and what your actual question is.
May
24
comment XenServer - Does the Hypervisor act as a router?
You could also create other vSwitches that aren't connected to a physical nic at all, to prevent broadcast or other traffic from being sent on the wire, although that clearly isn't the default configuration.
May
24
comment XenServer - Does the Hypervisor act as a router?
The default behavior is to create a virtual switch for each physical NIC in the system, and connect VMs to the first vSwitch that is created. If the VMs are just communicating with each other, and are on the same subnet, then the traffic won't leave this vSwitch (e.g. you could have the physical NIC disconnected and it would still work).
May
17
comment Reduce firewall rules by half - one iptables rule for tcp and udp
@davidkennedy85 If you really only care about HTTP, just drop the UDP rule and you've halved the rules. HTTP doesn't use UDP at all. (But make sure you actually know what is going on in your network before making changes you don't understand.) Also, you may wish to restrict it down to a small number of ports (the standard defaults for HTTP and HTTPS are 80 and 443 respectively), but again, that depends on what you are really trying to accomplish here. Overall, I'd recommend s.co.tt's answer as a better one.
May
17
comment Reduce firewall rules by half - one iptables rule for tcp and udp
@NathanC Yes, and removing the -p means "open up all protocols, not just tcp or udp", which is reckless if not dangerous.
May
17
comment Reduce firewall rules by half - one iptables rule for tcp and udp
@NathanC, I think the advice to open up ALL ports when the OP is asking how to halve his rules may lead to trouble, now or in the future.
May
15
comment What is the best way to copy 9TB of data to multiple external drives, sequentially?
Why does only one USB port dictate that you have to make the copy sequentially? Wouldn't a USB hub solve that problem? The answer to your actual question is below (I'm going with sysadmin1138's answer). Of course, depending on your data, you could just manually break it up into five groups and copy each one to a drive (e.g. copy user accounts a-e to drive one, f-k to drive two, and so on). And, depending on the data, if you back it up with compression it might actually all fit on one or two drives.
Apr
24
comment Multicast File Transfers
@SadeqDousti The advantages weren't so much about the speed difference, but more about usability, personal preferences, and our specific use case; which is why I didn't elaborate on them in the answer. You'll have to do your own evaluation to make a determination about which is better for your particular needs.
Apr
4
comment SFP+ vs CAT6a for 10GbE
+1 for the Direct Attach Cables (DACs). We love them in our environment, they are far cheaper, seem to have less issues with compatibility, and I've read they have slightly less latency than optics (although haven't tested this myself).
Feb
7
comment Why does traceroute take much longer than ping?
This does not make the round trip times any faster. It does however make it faster to return results on your screen, since it doesn't do the DNS lookups.