Posts by Alinator

21) Message boards : Web interfaces : Battery & User Active Preference (Message 12777)
Posted 27 Sep 2007 by Alinator
Post:
Also, just to make sure I read your previous post on the versioning correctly;

You're saying that the server part of a build will report itself in the CC message log as being whatever the version of the client portion was for the tarball it was built from,

Correct there.
even if the server part was actually older than that?

Not really; the server part doesn't officially have a version number, so you can't compare the version numbers ("older than that") :) Given how fast the server changes, the SVN revision works pretty well as a server version number.


OK... Sheesh, that's a fine how do you do. Why report anything at all (to the CC's) in that case! ;-)

I posted on the SAH and EAH NC fora to see if I can get the SVN rev # to post here and narrow things down a bit.

Alinator
22) Message boards : Web interfaces : Battery & User Active Preference (Message 12775)
Posted 27 Sep 2007 by Alinator
Post:
1.) All venues except the default will show the preference as 'No' regardless of whether you have set it to 'Yes'.

2.) If you have more than one extra venue set, if you change the preference on one of them to 'Yes', it will change it internally and relay it to the host on the next update. However, all the other named ones get reset to 'No'. In addition if you have set one of the named venues to work correctly (despite it showing otherwise), any change to any other pref on any venue ends up having the battery pref for the named venues reset to 'No'.

I don't own a laptop and I don't want to mess up with venues, so I can't test this myself. I'll post on the mailing lists pointing to this forum thread. Rytis (the maintainer of the web parts) isn't online right now, so you'll have to wait at least till tomorrow for a definite answer.


LOL...

Given what I've been seeing here, I can't say I blame you for that! ;-)

Anyway thanks for spreading the word around on the mail lists, and the tests are easily duplicated and repeatable.

Also, just to make sure I read your previous post on the versioning correctly;

You're saying that the server part of a build will report itself in the CC message log as being whatever the version of the client portion was for the tarball it was built from, even if the server part was actually older (or newer) than that?

Alinator
23) Message boards : Web interfaces : Server version numbers (Message 12771)
Posted 27 Sep 2007 by Alinator
Post:
Those are client version numbers; it's hard to know what server version matches with whatever client was available at the time. Instead, check the SVN revision shown on the project's server status page (it should be a 5-digit number).


Look on the CC message tab after a scheduler contact. It will show the server version, and in this case 511 on SAH nad 601 for EAH.

Alinator
24) Message boards : Web interfaces : Battery & User Active Preference (Message 12768)
Posted 27 Sep 2007 by Alinator
Post:
No, I am not talking about whether or not the CC's will execute the battery preference correctly or not, they do.

I'm talking about the website preference interface on server builds 5.11 and 6.01.

1.) All venues except the default will show the preference as 'No' regardless of whether you have set it to 'Yes'.

2.) If you have more than one extra venue set, if you change the preference on one of them to 'Yes', it will change it internally and relay it to the host on the next update. However, all the other named ones get reset to 'No'. In addition if you have set one of the named venues to work correctly (despite it showing otherwise), any change to any other pref on any venue ends up having the battery pref for the named venues reset to 'No'.

<edit> Looking back in my logbook, SB 5.09 worked correctly. I explicitly tested this when I noticed my notebook's battery get sacked quickly a couple of times when I knew for fact my prefs should have suspended work on batteries.

Since I routinely use SAH as my primary control project, I noticed they had upgraded to 5.11. I tried using EAH (which was still on 5.09 at the time) and it worked correctly, so I've using it as the primary since then. The other day my notebook battery got sacked again unexpectedly, so I checked and sure enough EAH had upgraded to 6.01, which is how I found the battery pref is still broken on more current versions.

Alinator
25) Message boards : Web interfaces : Battery & User Active Preference (Message 12753)
Posted 27 Sep 2007 by Alinator
Post:
Is the Battery preference ever going to get fixed for venues other than the default one?

It was broken on 5.11, and is still broken with 6.01.

Alinator
26) Message boards : BOINC client : 5.10.20 - uploading but not reporting (Message 12301)
Posted 4 Sep 2007 by Alinator
Post:
Yes, all the versions after 5.10.13 will sit on the report for a day, regardless of whether you have the CI set to less than that.

Personally, I think this is a bad idea. The fact of the matter is when you decouple and set the CI to a value less than the Cache setting, the user has explicitly told BOINC, "No, I DO NOT want you sitting on reports for longer than this amount of time" (among other things).

I'm tired of hearing the arguments of "Oh we have to do this in order to make better use of backend resources.... yada, yada, yada".

I've heard it suggested it's not a problem, because most hosts will come looking for work before the contact override times out. While that may be true, it makes the assumption that all hosts are fast, run 24/7, and/or have unfettered internet connection times. It also neglects to take into account a user may prefer to not have the fate of their completed work hinging on the integrity of an active working file like the client_state, which can get messed up for any number of reasons not even related to a BOINC malfunction per se.

The fact of the matter is that if a project has a problem with hosts 'pestering' it too frequently, the appropriate answer is for them to tell the Client during a contact session, "Oh BTW, I would like it if you would not contact me again about your work state situation for X amount of time".

It is not my hosts problem if a project chooses to 'overbook' their backend's capacity to deal with the volume of work they have in progress, or appropriate to have the CC override by default in all cases the manner which the people who donate their computing resources to a project choose to have their machines behave.

Alinator.
27) Message boards : BOINC client : Mac Boinc 5.8.17 not honoring preferences (Message 10080)
Posted 9 May 2007 by Alinator
Post:
OK, here's some clarification for what I posted on EAH regarding the "Do work while computer is in use" preference:

When working with the default preference mechanism:

The empty element <run_if_user_active/> toggles properly upon changes of the preference on the web site. IOW, I'm assuming here the presense or not of the empty element tag is the parameter that gets passed to the CC after parsing, and for 'Yes' it appears in the global_prefs xml file and for 'No' it does not.

In any event, 5.8.15 on Windows does not respond to it at all. The apps will continue to run whether it's there or not. I also tried using standard element syntax for it and that didn't work either.

When using the Local Overide mechanism:

Simply put, once you check the box to enable it from the Simple GUI it does not set up the <run_if_user_active>0</run_if_user_active> element when you select a timeout value for it initially.

If you then go back to the Simple GUI and select 'Run Always' it creates the element set to 1 like it should, but will not toggle it to 0 from the Simple GUI no matter what value you select for the timeout after that.

The bottom line is the only way to get it to work on Windows at all is to manually edit the global_prefs_override file.

Alinator
28) Message boards : The Lounge : What the heck is Sony up to? (Message 9627)
Posted 16 Apr 2007 by Alinator
Post:
Yeah right, after the CD debacle I'm going to trust SONY software on any DRM locked down platform.

They'll be skating down there first. ;-)

Alinator
29) Message boards : The Lounge : Question about Mail List Archives (Message 9607)
Posted 16 Apr 2007 by Alinator
Post:
What happened to the BOINC_Dev mail list archives?

All but this month and July '05 seem to have gone POOF.

Alinator
30) Message boards : BOINC Manager : 5.8.9 Issues on 9x (Message 8081)
Posted 8 Feb 2007 by Alinator
Post:
I'm looking for second source verification of some installer issues on 9x.

Specifically, when working from a full clean install:

1.) The installer inserts the startup at logon shortcut regardless of what you select during the installation. This has been an issue since version 4.

The Uninstaller is essentially completely broken when selecting remove for the uninstall option:

1.) It does not delete the files it is suppose to when uninstalling. I verified this by comparing the BOINC folder from a NT based system after running the remove option to the 9x system after doing the same.

2.) It does not update the registry or Start menu correctly to reflect the application has been removed.

3.) When exiting from the uninstall dialog it immediately restarts the Manager, which of course runs since it effectively has not been removed.

In essence the only thing it appeared to actually do was to remove itself from the Add/Remove Programs dialog.

Alinator
31) Message boards : BOINC client : Running BOINC as hidden (Message 6010)
Posted 14 Oct 2006 by Alinator
Post:
You can also give the CC the -detach option from the command line. You could use the switch in the login script you use to start BOINC and nothing will show on the taskbar or systray.

However as has been mentioned, for NT based Windows, a service install will allow BOINC to run even if no one is logged in with no fuss which usually is a big plus (assuming you're the administrator of course).

HTH,

Alinator

32) Message boards : BOINC Manager : Can't connect to Predictor@home (Message 5924)
Posted 5 Oct 2006 by Alinator
Post:
Yes, the project is down while they undergo a major overhaul.

However they aren't the most talkative project team around so don't be expecting to find very many updates on the status. ;-)

Alinator

33) Message boards : BOINC Manager : BOINC distribution under the GNU Public License (GPL) (Message 5923)
Posted 5 Oct 2006 by Alinator
Post:
@ Richard: Message at EAH received. Thanks. Took a while to find my BOINC Alpha Account Key! :-)

I've been rebuilding all my personal systems and LAN. Sheesh, I've accumulated so much "stuff" it takes a while to get all the ducks back in a row! ;-)

Guess that's why I don't do it too often anymore! LOL

@ Eric:

Thanks for taking the time to elaborate some on the state of BOINC policy as it stands. This has provided some additional food for thought.

I do admire the decision to go GPL with the software. Not only is the idea of BOINC pioneering in and of itself, but by releasing it as GPL it provides a vehicle to explore the uncharted waters of what the License means in a hopefully more, shall we say scientific rather than high stakes confrontational manner than would occur in the strictly commercial world.

Alinator
34) Message boards : BOINC client : 5.4.11 Bug (Message 5344)
Posted 16 Aug 2006 by Alinator
Post:
"CTRL-C" does not toggle between pause and resume properly (pauses but won't resume) when runniing from the command line.

This started after the transition from 5.2.x to 5.4.x.

Alinator
35) Message boards : BOINC client : Question about STD in 5x clients (Message 3276)
Posted 28 Feb 2006 by Alinator
Post:
It was my understanding that STD is normalized so that the smallest value is supposed to be zero, and tools like BV showed this was the case with 4x. However, I have noticed with the 5x clients BV indicates that the STD is being normalized to a mean of zero, like the LTD, and I was wondering what the story is with this.

FWIW, it doesn't seem to have a negative impact on operations, and task selection and switching behaviour appears to be what I would expect.

Alinator
36) Message boards : BOINC Manager : RC Password Suggestion (Message 2973)
Posted 6 Feb 2006 by Alinator
Post:
I long ago stopped using "dictionary" PW's and change them frequently.

How about making it so you have the option for the manager to remember the PW, or at least make it so you can paste it in.

This is the Win version I'm speaking of, but I assume this aggravation exists in the others as well.

In addition, I question the storage of the PW in the clear on the target host. Isn't there a decent open source crypto library you could use to protect the sensitive user account and PW info stored on the host?

Alinator


Previous 20

Copyright © 2024 University of California.
Permission is granted to copy, distribute and/or modify this document under the terms of the GNU Free Documentation License, Version 1.2 or any later version published by the Free Software Foundation.