- 29 Jul, 2004 7 commits
-
-
Mike Hibler authored
-
Mike Hibler authored
of affairs.
-
Mike Hibler authored
This is apparently a no-no and was happening at DETER.
-
Mike Hibler authored
-
Mike Hibler authored
It isn't needed on clients and it requires the linuxthreads package which is not otherwise needed.
-
Leigh B. Stoller authored
-
Leigh B. Stoller authored
-
- 28 Jul, 2004 7 commits
-
-
Russ Fish authored
I took it off when debugging popen streams on Windows; it shouldn't have been checked in.
-
Leigh B. Stoller authored
-
Leigh B. Stoller authored
to be reused if the DB is dropped and recreated, since when that happens, auto_increment history is lost and it will go back to using the latest highest index in the table. Usually not a problem, but since we cross index three other tables using the experiment index, this causes quite a bit of grief. So, my solution is to do my own auto_increment using the experiment_stats table (locked of course), which we never delete entries from without deleting all entries from the other cross referenced tables. DBQueryFatal("select MAX(exptidx) from experiment_stats"); I also added a sanity check to make sure the new index is not currently in use in any of the tables. I also cleaned up the error path when something goes wrong.
-
Russ Fish authored
-
Russ Fish authored
Add a cross-reference from the Windows installation section to the release notes section on setting up SSH or PuTTY/plink.
-
Leigh B. Stoller authored
-
Leigh B. Stoller authored
NFS, convert to using a proxy that runs on ops, which does the copying locally.
-
- 27 Jul, 2004 5 commits
-
-
Kirk Webb authored
Wrote new abstract and outline for WORLDS paper based on today's paper meeting.
-
Timothy Stack authored
webfeedback to talk to the newer version of canaryd. Add feedback "estimate" stuff so that if we have no data (because of an overloaded node) to work with, we can make some sort of "reasonable" guesstimate on every iteration.
-
Timothy Stack authored
than jamming more stuff into the old one. Most of the code came from the previous version of canaryd, the cpu broker (process accounting), and the janosvm (network interface accounting). Its missing some features of the old one, but those can be incorporated without too much trouble. Changes: Designed to permanently run on the pnodes: it waits for START events before it begins recording. However, I haven't done the work necessary to have it always startup on the pnodes. No more exec'ing: process stuff is taken from "/proc", and network interface stats are pulled from getifaddrs(3). Fixed some minor bugs: A typo caused the real-time priority to not be set, use setitimer instead of sleep to get more accurate spacing between samples.
-
Leigh B. Stoller authored
return instead.
-
Leigh B. Stoller authored
at the same time you clear the load address. admin people only.
-
- 26 Jul, 2004 7 commits
-
-
Timothy Stack authored
instead of just their names. Enhance experiment.statewait() to accept multiple states to wait for, or, to wait for any state change. Add experiment.thumbnail() method so the topology can be displayed in NetBuild.
-
Timothy Stack authored
event system from crashing when multiple EventClients are created.
-
Timothy Stack authored
-
Leigh B. Stoller authored
Fix rather serious bug with how ranges interact with failed swapmod. Actually, the bug is mostly in how a failed swapmod (or swapout) results in missing testbed_stats records. I'm working on fixes for that.
-
Leigh B. Stoller authored
experiment is swapped or 2) the experiment is completely terminated. In these case, lets put explicit swapout/destroy events into testbed_stats so that the record is not confused by experiments that appear to start when they are still running. This really throws off the summary stats web page!
-
Leigh B. Stoller authored
-
Leigh B. Stoller authored
-
- 23 Jul, 2004 3 commits
-
-
Jay Lepreau authored
Ram into intro for the moment.
-
Jay Lepreau authored
This is just a temp thing not to to miss some possibly useful text.
-
Jay Lepreau authored
-
- 22 Jul, 2004 1 commit
-
-
Leigh B. Stoller authored
goes whacky.
-
- 21 Jul, 2004 3 commits
-
-
Russ Fish authored
Address some feedback comments, add more TOC, and add an extended example exploring the 2440 experiment graph and using the root glob expression option to experiment reading.
-
Russ Fish authored
-
Leigh B. Stoller authored
-
- 20 Jul, 2004 6 commits
-
-
Russ Fish authored
-
Russ Fish authored
I will have to fix the bug in PyOpenGL argument passing to get picking working right. Meanwhile, I put in a simple work-around hack: when you pick a graph edge in the spanning tree displayed by HyperViewer, the node it leads to is picked instead. (The edge has a negative integer ID, while the node it leads to is positive.) This also gives the ability to "zip" along edges of the graph, which I kind of like. It's not a very good substitute for picking nodes, but might be worth keeping even after node picking works. We can also define our own semantics for picking edges in the graph...
-
Russ Fish authored
-
Shashi Guruprasad authored
an OS, the code to check for the presence of non-BSD nodes would fail. Fixed the check so that we consider a PC to be a non-BSD node if we have no information about what default OS is gonna be loaded on to a PC after mapping. Lunquan Li of WormE project triggered this problem. BUGFIX
-
David G Andersen authored
IPOD packet, now you can. (Note that having ipod installed setuid is already a moderate risk, so i don't think this makes it too much worse)
-
Russ Fish authored
-
- 19 Jul, 2004 1 commit
-
-
Russ Fish authored
-