| Index | Recent Threads | Unanswered Threads | Who's Active | Guidelines | Search |
 |
World Community Grid Forums
Category: Completed Research
Forum: FightAIDS@Home
Thread: No tasks Available for PCs[RESOLVED] |
| No member browsing this thread |
|
Thread Status: Active Total posts in this thread: 18
|
|
| Author |
|
|
Former Member
Cruncher Joined: May 22, 2018 Post Count: 0 Status: Offline |
They seem to have fixed it, now got lots of FAAH
 |
||
|
|
acpartsman
Veteran Cruncher Martinsville VA, USA Joined: May 6, 2007 Post Count: 943 Status: Offline Project Badges: 
|
Np tasks available on Androids running newest release, doesn't really matter though since ALL WUs go into Computation Error anyway.  Androids running version from last September still receive and process WUs flawlessly.  As I've said before, roll back the project to last Septembers version, it obviously had no issues. My highlighting, not the first time you tell us of this riddle and still not comprehended, by me at least. Possibly your agent has now been updated since then? Far as can be seen the android fahv science application was but for yesterdays 7.16 downloaded version not replaced since september launch with version 7.06 and to emphasize, you have absolutely NO control over the update of the science application. It was noted as being fully automatic, the question thus being what agent/science version of last year was running flawlessly? We all like to know. Also, you can always go back to any agent apk you had which can be downloaded from the berkeley /dl archive if the newer boinc for android release are the cause. The Android phone that keeps giving Computation Errors is my best phone, one that has been crunching since the beginning of the Android projects. And now even with the new update last night it still gives computation errors (when it gets WUs) The phone that continues to run flawlessly is a much lesser phone and is set to not automatically update, therefore it is running on the version that was current in September. As for loading any version I want from that list you mentioned.....that does NOT work because it has to be installed through PlayStore and those older versions are not available there. Loading through BOINC site only results in download error, unable to install as there is no program on the phone to decode that file. As for the version that is running on the other phone, I can't tell you because that version doesn't ID itself.
One drop raises the sea.
 |
||
|
|
Former Member
Cruncher Joined: May 22, 2018 Post Count: 0 Status: Offline |
"The phone that continues to run flawlessly is a much lesser phone and is set to not automatically update, therefore it is running on the version that was current in September."
Been loading apps outside of the playstore, needing a tweak in the settings > security > unknown sources. When done, reversing that so nothing more can slip in should it be pushed. Also at individual application level in the playstore been ticking which app is allowed to auto update. This gives control as to seeing which ones want to sneak a new release. Google products are the most prolific in that respect. No about box somewhere in there or in the playstore my app screens? Click boinc to open it's full info screen in the playstore, which in mine tells I'm supposedly having 7.3.7. Not so as outside I've updated to 7.3.12. Think it would be really good to know the 'old' version of september 2013 as it could give developers a lead as to what change might be the cause. All those 195 errors are rather distressing if you have a device generating those. Not here with the new 7.16, but instead a few signal 11's as in overloaded. |
||
|
|
acpartsman
Veteran Cruncher Martinsville VA, USA Joined: May 6, 2007 Post Count: 943 Status: Offline Project Badges:
|
Finally received a WU this morning, Computation Error showed within an hour. I've had it. Discontinuing crunching, tired of burning up my phone for nothing. No usable work in over 7days.
---------------------------------------- Maybe when enough people complain something will get done.
One drop raises the sea.
|
|||
|
||||
|
uplinger
Former World Community Grid Tech Joined: May 23, 2005 Post Count: 3952 Status: Offline Project Badges: 
|
We are looking into the issue. Can you try to decrease the number of cores used on the device to 1 and see if that helps with the issue? Out of the 7 different physical test devices we have only 1 of them shows this behavior and it's not all the time which makes debugging very slow. We are trying to resolve this issue as quickly as possible.
Thanks, -Uplinger |
||
|
|
Former Member
Cruncher Joined: May 22, 2018 Post Count: 0 Status: Offline |
Another strange thing is to keep seeing wingmen on android with user aborts that have version 7.20, when the android version is 7.16, or? Are you mixing Linux with Android? Do they not go in separate homogeneous redundancy groups? Not helping you to keep reported fails below a number that does not lead to otherwise valid returns stuck in pending validation to go too late, 9? Of that one I see a task now this moment that went 9 times error, and one waiting 'in progress' with stopped circulation. Cooked, fried and sauteed the situation feels to be, the one who is succeeding to compute the 10th is 100 percent sure going to waste time. Given I've had several 22 hour runs, not small. Is a server abort message sent in these instances?
----------------------------------------[Edit 1 times, last edit by Former Member at Mar 31, 2014 3:30:57 PM] |
||
|
|
acpartsman
Veteran Cruncher Martinsville VA, USA Joined: May 6, 2007 Post Count: 943 Status: Offline Project Badges:
|
We are looking into the issue. Can you try to decrease the number of cores used on the device to 1 and see if that helps with the issue? Out of the 7 different physical test devices we have only 1 of them shows this behavior and it's not all the time which makes debugging very slow. We are trying to resolve this issue as quickly as possible. Thanks, -Uplinger My phone is a single core device (Samsung SCH-I405U on v2.3.6). Still continue to get a 90%and higher computation error rate. I've tried setting it to process only on charger and run always down to 30% battery, nothing makes a difference. Computation errors occur anywhere between immediately after download and almost complete.
One drop raises the sea.
|
||
|
|
uplinger
Former World Community Grid Tech Joined: May 23, 2005 Post Count: 3952 Status: Offline Project Badges:
|
lavaflow,
We are not mixing the application versions. What you are probably seeing is that someone downloads a linux work unit and their machine is set to abort immediately. This gets reported back to the server, since there are no versions of this work unit as success or in progress to a certain computer class it gets opened back up to all classes. acpartsman, I am sorry for the problems you are having, we are working on solving these issues, but it's a slow process due to recreating the issue. Again, we are working towards a fix. Thanks, -Uplinger |
||
|
|
|