GitLab

named user to "bind"

 The nfsd script will try to kill all nfsd processes, but once
 the master is killed, it kills all children itself.  Thus the
 nfsd script may fail to kill the children if they are already
 dead.  "forcestop" instead of "stop" tells the script to ignore errors.

No idea why rpc exchanges started failing after updating boss/ops to
FreeBSD 6.1 or, for that matter, why this really fixes the problem.
Monitoring the exchanges between client and the RPC serial line didn't
show any rhyme or reason why some exchanges elicited "Input error" and
others didn't.  But I noticed that every \r we send to the RPC gets
echoed back as \r\r\n (and no, it isn't the serial line CR<->NL mapping).
So I reduced the number of \r (actually \n) we send it and things got better.

This "fix" has been running every 5 minutes for the last hour without
failure, so I declare it: An Improvement.

get this right since the lexical environment inside an eval inside a
package is not what I thought it would be!

mysql injection attack. Been there for years. Amazing.

The adminon/off bit is now kept in the login table instead of the
users table so that you can be reddot on a per-browser (login) basis.

Fix the last fix...

Comment out a dubious piece of code that "adjusted" the recorded runtime
of iperf if the packet loss was greater than 10%.  I think this code was
another way of dealing with the lost/retransmitted FIN problem that has
been addressed with a new iperf option.

Fix up how vnode_setup wait()s for it's children to deal with a change in how
perl deals with interrupted system calls.  What the script was doing
previsiously was iffy anyway.

on /tmp which is now on /q.

environment from the environment strings contained in the event.
This is currently used when starting a new template run so that new
parameters can be passed along.

on delay nodes.  E.g., a delay of 5ms is split up as 2.5ms for every node in
the LAN, and we round that to 3ms before handing to dummynet.

This will produce linktest errors.

Add a KB "Testbed Operations" section for admins (red-dot) or foreign_admins (from the testbed-admins list.)

Something subtle changed with the upgrade to FreeBSD 6.1,
maybe something to do with perl buffering?

system version of bind instead of the port. This includes using
the startup file in /etc/rc.d.

Only show KB "Testbed Operations" section for admins (red-dot) or foreign_admins (from the testbed-admins list.)

Make sure suidperl is not setuid on FreeBSD 6 or later.

mess I created.

Only show KB "Testbed Operations" section for admins (red-dot) or foreign_admins (from the testbed-admins list.)

Going to a file instead of the DB (too much hassle). This needs more
work as I am still unclear as to how timelines, sequences, etc fit
into the event stream.

The file is placed in the archive and committed with each Run and Swapout.

At present, only template derived experiments do this ...