GitLab

only pid, to pid/gid like most other things in the testbed. Also add a
"global" slot to denote images that are globally available to all
projects (system images). The older "shared" attribute is now used to
denote images that are shared within a project (available to all
subgroups in the project). The migration path for existing DBs is
given in the migrate file. Be sure to run those commands on an
existing testbed or things will break!

www/newimageid, www/newimageid_ez: A bunch of changes for
shared/global attributes. Added a group menu to the form so users can
create images in subgroups. Beefed up the Java code that constructs
the path name to use the gid, shared, and global attributes of the
form to give the user the best possible path that we can. Improved the
pathname checking code so that we do not allow just any old path in
case the user elects to disregard the path we carefully constructed
for them. Also check the proj/group membership, and setup defaults for
users that have permission in just one pid/gid to create images.

libdb.in: Changed permission check in TBImageIDAccessCheck() to
reflect shared/global attribute changes.

os_load: Get rid of test that checked path of the image. The path
checking is done in the web interface anyway, so why duplicate in 4
places. Other minor changes reflecting shared->global name change.
Also note that images can come from the group directory now.

create_image: Get rid of test that checked path of the image. The path
checking is done in the web interface anyway, so why duplicate in 4
places. Also note that images can come from the group directory now.

www/dbdefs: Changed permission check in TBImageIDAccessCheck() to
reflect shared/global attribute changes.

www/showimageid_list, www/showstuff: Minor global/shared attribute
changes.

www/menu: Change osids/imageids pointer to point to the image list,
not the osid list. This is more reasonable for mere users who have
access to the EZ form, and thus never really need to concern
themselves with osids.

www/editimageid: Add proper pathname checking. There were no checks at
all before!

 - Added a bunch of functions to get time of last activity, hours idle,
   and staleness for nodes and experiments.
 - Fixed the showexp_list page to use the new idle measures, including
   idle view. Idle view now runs in about 1.3 seconds instead of 13.
 - Changed all the "show" functions for nodes and expts to show idleness.

More changes will be coming, but I wanted to commit this before I install
it for real.

when a new LAN is created, we make a new plink. Since assign is only
halfway OO, there is no constuctor for a plink to set the fixends
flag to something sensible.

Turned back on FIX_PLINK_ENDPOINTS, and 'touch' score.cc so that it
gets rebuilt.

   - Performing an approval action (postpone/approve/deny/nuke) on a
     user in a subgroup results in an implicit version of that same
     approval action on the user in the
     default group (for approve, $trust='user'.)

   - Tidied up a table in approveuser_form.php3,
     but form is otherwise unchanged.

the newimageid_explain front page, since I had actually moved most of
that text into the tutorial a couple of months ago. I moved the rest
of the text over, and changed the links to point to the short form.
Added a link at the top of the short form pointing to the tutorial
section, and for admin types, a link to the long form.

Account to make it more clear.

Bad idea. Use an auto_incrementing idx field instead.

Fixed bug where after running out of retry attempts,
swapout was only doing the 'retry' version of cleanup,
when it should have been doing a full cleanup.

and a query returns a null where it ain't supposed to!

by assign_wrapper to help control emulated links. When set for an
emulated link, assign uses the BW spec for mapping the links (onto
phys links) but does not insert a linkdelay to traffic shape it
(unless the link is shaped in some other manner, like a delay or plr,
in which case a linkdelay is inserted, but with no bw cap). This might
be temporary syntax, expecially if we cannot make emulated links work
right!

kernel since its hacks are so untested.

links, no lans. The whole lan thing needs work.

Add better timeout handling code to tmcc, which watches for progress
instead of just dumping after the timeout. This lets really slow
connections proceed okay, but still timeout out if nothing happens at
all.

This new tmcc has been installed into the sup trees for WIDE and RON.

idle but someone logged in sometime today, it meant that someone had
logged in since the most recent slothd report (up to an hour ago). Now
that our time of idleness is much lower than 24 hours, it is easy for
someone to have logged in today and still have it be idle. So remove the
check.

This means that you should wait at least an hour from the time you send a
message before you swap it out, to make sure that your data was correct
when you sent the message. Not optimal, but fixed with the new slothd that
goes in this week.

it was used outside the NTFS code.  imagezip builds correctly once again.

on some plinks, an apparent memory corruption bug.

under pressure; should stick to watching cartoons instead.

Scaling and speed now depends primarily on os_load (and indirectly,
node_reboot). The time a batch spends in the reload_daemon code appears to
be <1s per node now, instead of taking 30s per node to grab, setup, and
reboot.

Also, finally remove the "obsolete section" that's been sitting in there
for a long time. This was the part that did netdisk reloads, and has
already been neutered out of the code path for several months at least.

of emulated and linkdelays on a per-link/lan basis (although you
cannot make a lan emulated at this time).

	tb-set-link-uselinkdelay $link0 1
	tb-set-lan-uselinkdelay $lan0 1
	tb-set-link-emulated $link0 1

Added slots to the virt_lans table, used in assign_wrapper to control
these features on a per link/lan basis. Note, the global options are
still available to prefer/force linkdelays for an entire experiment.
When setting uselinkdelays on a per link/basis, if the link would not
otherwise be shaped, you are requesting that a linkdelay be inserted
anyway (in otherwords, its the equiv of tb-set-forcelinkdelays on an
individual basis).

have gone active again. If you gave it a higher time than the default, and
an idle expt hadn't been idle that long, it would clear the counter when
it shouldn't, making idle expts look like they may have gone active again.
(It didn't ever change any slothd data or cause any activity, you'd just
"forget" that you'd already been sending swap requests. And it only had an
effect when you picked times _higher_ than the default idle time.)

reason is to keep things moving during the high SOSP demand, but it may
stay there or somewhere similar when we deploy the new slothd next week.

parameters). Install target update as well.

Several things have changed; here is an outline:

 - "Thresholds" added to slothd
	These are checked at each iteration against current idle data
        to determine whether or not to set bits in the "active" vector.
	- LAST_TTY_ACT (now) > LAST_TTY_ACT (previous)
        - LOAD > <provided value>
        - EXPER_PKT_COUNT(now) - EXPER_PKT_COUNT(prev) > <provided value>
	- CNTRL_PKT_COUNT(now) - CNTRL_PKT_COUNT(prev) > <provided value>
	- Values passed in via command line: do ./slothd -h for help
 - "Aggressive" mode added to slothd
	When a regular interval's worth of time has passed, and no activity
        has been noted, slothd goes into "aggressive" mode.  In this mode,
	slothd changes its sleep interval, checking for activity much more
	frequently (still reporting every <reg_interval>) until activity
	is noted.  This is based on the threshold logic above.  When activity
	is seen, slothd sends a report, and goes back to "regular" reporting.
	- params passed via command line for regular, and aggressive intervals.
 - sdcollectd updates data in new "node_activity" table
	- Updates timestamps based on activity bits in received slothd packets
	- No new rows are added.
 - Option added to sdcollectd to insert into "old" node_idlestats and
	iface_counters tables
   	- This is not the default (-o)
 - New sdcollectd still accepts old client reports
	- but does not update the node_activity table with them.
 - Misc.
	- Code cleanup
	- Scan "active terminals only" option & code removed from slothd
	- Always send report immediately on startup (no more option)
	- Randomize time of second report to mitigate synchronization
	- Send report on shutdown (receipt of INT, TERM, or QUIT)

when doing emulated links so that ports do not end up in more than
one vlan.

Minor changes: Updated code to work with the new libntfs library, and made the NTFS section of the code adhere to Leigh's 80 column decree.

virtual links.

There will need to be a way to specify this on a per-vlink basis
for jails.

* Make php.ini file

* Edit ssh_config to prefer protocol version 1

* Make /usr/testbed/tftpboot, and link /tftpboot to it

* Make mountpoints for places we mount ops

* Make sure that the DNS names for ops resolve before we put them
  into /etc/fstab

* Chase emulab-boss port version