Email addresses
egrep -o ‘[A-Za-z0-9._]+@[A-Za-z0-9.]+\.[a-zA-Z]{2,4}’ filename.txt
URLS
egrep -o “http://[a-zA-Z0-9.]+\.[a-zA-Z]{2,3}” filename.txt
Email addresses
egrep -o ‘[A-Za-z0-9._]+@[A-Za-z0-9.]+\.[a-zA-Z]{2,4}’ filename.txt
URLS
egrep -o “http://[a-zA-Z0-9.]+\.[a-zA-Z]{2,3}” filename.txt
This weekend I started playing around with TCP Tuning and playing around with the different windows sizes. This was brought up upon having a strange issue at work with a couple of SLES11.1 boxes experiencing some very sluggish network transfers.  Figured rather than make the situation worse by attempting to modify settings I’m really not familiar with , figured I’d play in my test lab at home where breaking something isn’t going to bother anyone except me. Anyhow, long story short,  I was shocked to see some of my scp jobs that I was using to copy 3GB files around where also very sluggish after about ~1500MB. I even got the same network “stalled” warning during the transfer a couple of times.
After attempting to modify the TCP window sizes and tried different settings, I still was having the same issues.  Usually starts around ~1GB-2GB using different values. Turns out, SCP I guess has it’s own internal buffers built in, so when you are scping small files you never notice this but for larger files, they say that the host can’t keep up with encrypting the file after a certain threshold. There is a patch for sshd or the workaround is using scp -z which doesn’t encrypt the file .  Your login info is still encrypted but not the file your copying. Using -z isn’t an acceptable solution.
More info on this can be found here..