Home / user-43511

Steve Kehlet's questions

Martin Hope
Steve Kehlet
Asked: 2014-04-05 10:12:33 +0800 CST
  • 1

I want to show zero-byte TCP keepalives sent and received. While this could be generally useful, in this particular case I want to prove that a change I made to the JDBC driver (adding tcpKeepAlive=true) is causing the keepalives to be sent periodically (at the interval I've already configured in /proc/sys/net/ipv4/tcp_keepalive_time, and I know is working). They look like this:

21:35:00.150604 IP 1.2.3.4.53019 > 5.6.7.8.postgres: . ack 635465 win 772 <nop,nop,timestamp 1666835314 1722393170>

But I'd like to figure out a filter so it doesn't show all traffic. I found ip[2:2] which allows you to filter by the total packet length. Is the length of a zero-byte ACK constant?

# not it, but maybe close
tcpdump -nn 'ip[2:2] = 50'

Thanks!

tcpdump
Martin Hope
Steve Kehlet
Asked: 2012-06-09 09:57:55 +0800 CST
  • 3

I'm investigating some application issues on a Linux system and noticed this around the time of the most recent problem, but not close enough to be definitively related:

Jun  7 17:51:49 localhost kernel: shrink_slab: nr=-155456000
Jun  7 17:51:49 localhost kernel: shrink_slab: nr=-157859400
Jun  7 17:51:49 localhost kernel: shrink_slab: nr=-157833400
Jun  8 06:20:23 localhost kernel: shrink_slab: nr=-284172800

What is this? Googling isn't helping, I get that it's something with VM, but why is it suddenly showing up in /var/log/messages?

Edit: I take it it's showing this because the number of pages it could free ("nr") is negative. Which is presumably a bad thing. I found patches for newer kernels that display a more useful message to this effect. Still not sure what this means though, or what I need to do about it.

linux kernel virtual-memory
Martin Hope
Steve Kehlet
Asked: 2011-05-27 10:16:08 +0800 CST
  • 5

In my linux startup scripts, when I start a process like (openvpn used for this example, but the question is general to any process);

openvpn --config /etc/myserver.conf

what is the best way to find it and be 100% certain it's the right process and kill it in the stop section? I normally use stuff like:

pid=$(ps -efww | grep -v grep | grep openvpn | grep /etc/myserver.conf | awk '{print $2}')

Sure it works nearly all of the time, but sometimes there are issues with accidentally matching processes with nearly the same name (e.g. myserver.conf-new), so I'm looking for a better approach.

  • Some processes have a way to store the pid somewhere, that's fine, but generally I'm skeptical of killing processes solely based on a pid sitting in a file somewhere.
  • Solaris has projects, it's not all roses in my limited experience because you have to set up /etc/projects first, but it does make it easy to tag and later find processes.
  • Maybe using the environment, like setting an environment variable like (MYID=myserver) and then looking for it with ps e -ef | grep MYID=myserver? Still may suffer the same problem with accidental matching.

I wish there were something easy like:

launch --tag myserver openvpn --config /etc/myserver.conf

and

pgrep --tag myserver
linux startup-scripts