GitLab

experiment info page redraw every 30 seconds when the exp state is in
the activating or swapping state. Eventually will stop when the state
changes.

users know about email and password stuff.

in. This was mostly to support the modal change in the home link, but
since people do not like that, this stuff is probably going to get
tossed at some point.

information. Also change order of info so that Profile is last.

not supported in the front end.

configuration commands in /etc/testbed/rc.delay.

This script rolls the apache logs in a friendly manner by gracefully restarting apache after its logs have been "mv'ed", causing it to reopen new logs, while keeping existing connections alive.  Plenty of saftey checks to make sure nothing gets lost/stomped.

tired of clicking through to find my own info and experiments.

echoed to stderr.

will roll the base log if it has been mv'ed out of the way. This log
stores the queries, and can get rather large very quickly.

in human form. This should save some space.

did the lanlink change. assign_wrapper does not really need any
changes at all, but I dropped a comment in. The only real change is
needed in the client side to convert a floating point from tmcd, into
an int to give to ipfw. Rather cheesy too, given that Perl does not
have a real conversion function (int() truncates).

collection daemon, and its peer query program (to be used by cricket, and
elsewhere if desired, to get sensor readings for nodes which have reported
in).

original healthd-0.6.5 codebase plus my linux port and a 'push' mode for
central, unsolicited reporting.  The code for the healthd collection daemon
is kept elsewhere in the 'hmcd' subdirectory under 'sensors'.  Also included
is a 'quiet' switch (-q) for healthd (no syslog entries for sensor conditions).

extra unix groups (unixgroup_membership) for special local users that
need more groups than just their project membership (ie: flux, wheel,
etc). In mkacct-ctrl, no longer use the admin bit to determine extra
groups (which were hardwired in), but get the extra group list from
the DB. This applies to accounts on boss/users; experimental nodes
still use the admin bit (via tmcd) to get wheel added to the group
set. Might be worth doing at some point.

capability. New DB field in the users table (pswd_expires) which is a
date field that initially gets set to one year after the user account
is created. When the password is changed via the web form, it gets
bumped 1 more year into the future *unless* the current uid is
different from the target_uid (ie: you are changing a password for
someone else). In that case, the expiration is set to the current
date, which forces the target user to change his password next time he
logs in. I've changed the menu/auth code to look for password
expiration, and when expired the menu options contain just a single
option to change the password. All other https pages will fail with a
password expired message. Normal text pages will work of course.

public bit. Moved the delete button to the project listing page since
it was wasting space and its rarely if ever used.

reloading experiment. Not particularly elegant, but a better solution should
come with the event system.

the exp/tbdata directory too.

sorted view, much like the public listing, with a section for active
and a section for inactive. The alternate view is accessed via a link
at the top of the page that allows you to switch between them.

sched.h. Renamed sched.c to event-sched.c to match header rename.

less.

so that admins can vis any experiment (that is, have access to the
exp directory in the /proj/pid tree). Eventually we want a standalone
top file generator.

lot more can be done with this, of course.

in the name. I changed that, but in the meantime search for top file.
Also redirect stderr to /dev/null when generating gif for the web page
since it will mess up the gif image.

the virtual topology and the physical topology. See equiv changes to
constants in ptopgen. The interface_types table is also changed to
reflect Kbps instead of Mbps.

distinguish that in the update user information page, and that makes
no sense. We should have that info for all users.

good to leak to the user, and it also messes up fpassthru in php.