Windows S5R2 App 4.33 available for Beta Test
Bernd Machenschalk
Moderator
Administrator
Joined: 15 Oct 04
Posts: 4312
Credit: 250214833
RAC: 35613
26 Jul 2007 15:51:33 UTC
Topic 192998
(moderation:
A new Windows App is available from our Beta Test Page.
This is our new "release candidate" that, if the Beta test doesn't reveal any serious problems, will be made "official" in the next few days. It should then fix at least the serious "Access Violations" in the "houghmap" module.
There is not much change to the 4.32 App, except for even more paranoid error checking and reporting to track down the remaining problems (including the fancy "empty or vanished checkpoint file" (resulting in an exit code 10)).
Please test and report.
BM
BM
Windows S5R2 App 4.33 available for Beta Test
)
Howdy,
are the following errors expectable:
7/26/2007 8:38:13 PM|Einstein@Home|Found app_info.xml; using anonymous platform
7/26/2007 8:38:13 PM||[error] No URL for file transfer of einstein_S5R2_4.32_windows_intelx86.exe
7/26/2007 8:38:13 PM|Einstein@Home|[error] Can't initialize file transfer for einstein_S5R2_4.32_windows_intelx86.exe
( I only habe now the 4.33 exe in the einstein project directory .. =
Also there seems to be maybe one extra character in the info.xml:
( the backtic .. )
file_info>
einstein_S5R2_4.33_windows_intelx86.exe
`
-- Lundi --
Thanks for spotting
)
Thanks for spotting this!
I fixed the app_info.xml and updated the package (and the md5 sum).
Sorry for this.
BM
BM
No problem at all .. When we
)
No problem at all .. When we talk about "beta" I take granted that there are "challenges" ..
But I still get the same errors:
26/07/2007 22:10:43||Starting BOINC client version 5.10.13 for windows_intelx86
26/07/2007 22:10:43||log flags: task, file_xfer, sched_ops
26/07/2007 22:10:43||Libraries: libcurl/7.16.1 OpenSSL/0.9.8e zlib/1.2.3
26/07/2007 22:10:43||Data directory: E:\\boinc
26/07/2007 22:10:43|Einstein@Home|Found app_info.xml; using anonymous platform
26/07/2007 22:10:43||[error] No URL for file transfer of einstein_S5R2_4.32_windows_intelx86.exe
26/07/2007 22:10:43|Einstein@Home|[error] Can't initialize file transfer for einstein_S5R2_4.32_windows_intelx86.exe
( I downloaded once more the 4.33 binary + the xml-file to be sure that they were "the right ones" )
What wonders me is the the following text block:
einstein_S5R2_4.33_windows_intelx86.exe
Without knowing anything I wonder where the pair for the is ( should it be before that? line like : ?! )
-- Lundi ---
I have downloaded/installed
)
I have downloaded/installed the revised 4.33 zip, and it ran without problems: the current (monster) WU started with 4.30, did a substantial amount of work with 4.32, and picked up cleanly with 4.33.
I did an 'eyeball' check of the app_info files, and apart from the extra version 433 segment, the only difference that I could see was the backtick, which was present in all the 4.28, 4.30, and 4.32 versions! Can't have been too much of a problem, then.
The only thing I can't comment on is validation - can't get a wingman to stay the course for this box.
Edit - I've been round the shrubbery, and updated 5 different versions of Windows - from 98SE to Vista. All restarted without a hitch. If you want to check results, all computers on my account are now running 4.33 (though the current WUs all started on something earlier).
RE: No problem at all ..
)
Not a problem. That syntax is what's known as the empty element tag.
In this case the presense of the tag is the parameter which is being looked for.
Alinator
The beta page instructions
)
The beta page instructions should be updated, because it still says
And it is not necessary anymore, because now you stop BOINC without any additional step, unzip the files, and when your restart BOINC the last application (4.33 in this case) replaces the old one (allthough BOINC keep showing the last application name in the cache's WU). I suppose app_info.xml does this instant upgrade.
However the advice is still
)
However the advice is still germane as it is still the safest way to go about making modifications to an active project, regardless of whether other methods work most of the time. Although the aborting unstarted tasks is usually not neccesary.
Alinator
This result was started with
)
This result was started with v4.32. I installed v4.33 at roughly the 70% mark. However, I don't see any indication of switching apps in its stderr_txt (as had been the case with earlier hybrids).
http://einsteinathome.org/task/85862719
RE: This result was started
)
You did restart BOINC , though?? :-)
In the results view, you will see only the last few kilo bytes of the stderr.txt file that is written while E@H runs (very roughly covering about 20% of the run), so the point where the apps switched may have "scrolled" out of this window. If the point of switching is sufficiently close to end, you will see something like this (from http://einsteinathome.org/task/85901682):
69331, 69332, 69333, 69334, 69335, 69336, 69337, 69338, 69339, 69340, 69341, 69342, 69343, 69344, 69345, 69346, c
69347, 69348, 69349, 69350, 69351, 69352, 69353, 69354, 69355, 69356, 69357, 69358, 69359, 69360, 69361, 69362, c
69363, 69364, 69365, 69366, 69367, 69368, 69369, 69370, 69371, 69372, 69373, 69374, 69375, 69376, 69377, c
69378, 69379, 69380,
2007-07-27 18:32:32.1599 [normal]: Start of BOINC application 'projects/einstein.phys.uwm.edu/einstein_S5R2_4.33_windows_intelx86.exe'.
2007-07-27 18:32:33.3917 [debug]: Reading SFTs and setting up stacks ... done
2007-07-27 18:32:48.2230 [debug]: Found checkpoint - reading...
2007-07-27 18:32:48.2230 [debug]: Read checkpoint - reading previous output...
2007-07-27 18:32:48.6436 [debug]: Read exactly 1034193 == maxbytes from Fstat-file, that's enough.
2007-07-27 18:32:48.6436 [debug]: DEBUG: read_fstat_toplist_from_fp() returned 1034193
2007-07-27 18:32:48.6436 [debug]: Total skypoints = 70358. Progress: 69378,
$Revision: 1.45 $ OPT:0 SCV:2, SCTRIM:8
69379, 69380, 69381, 69382, 69383, 69384, 69385, 69386, 69387, 69388, c
69389, 69390, 69391, 69392, 69393, 69394, 69395, 69396, 69397, 69398, 69399, 69400, 69401, 69402, c
69403, 69404, 69405, 69406, 69407, 69408, 69409, 69410, 69411, 69412, 69413, 69414, 69415, 69416, 69417, 69418, c
69419, 69420, 69421, 69422, 69423, 69424, 69425, 69426, 69427, 69428, 69429, 69430, 69431, 69432, 69433, 69434, c
69435, 69436, 69437, 69438, 69439, 69440, 69441, 69442, 69443, 69444, 69445, 69446, 69447, 69448, 69449, c
As you can see, the app version 4.33 continued the task after I re-started BOINc, but the result (see url above) lists the app version as 4.31, the version the task was started with.
So I would not be concerned, most likely everything is fine with your installation.
CU
BRM
RE: You did restart BOINC,
)
I exited BOINC, dragged the v4.33 files to my Einstein folder and then restarted BOINC. So, I guess the message got "scrolled" out. However, "the last 20%" rule of thumb you refer to can't be very accurate. The result I cited in this message showed the switchover and it occurred fairly early in its processing (roughly 15% from the start).