Hi, I use Win XP 64 bit.
I instaled Boinc client 5.10.7 version for 64 bit platform. And I find in log:
2007-07-03 22:08:08 [climateprediction.net] Sending scheduler request: Requested by user
2007-07-03 22:08:08 [climateprediction.net] (not requesting new work or reporting completed tasks)
2007-07-03 22:09:16 [climateprediction.net] Scheduler RPC succeeded [server version 509]
2007-07-03 22:09:16 [climateprediction.net] Message from server: platform \'windows_x86_64\' not found
2007-07-03 22:09:16 [climateprediction.net] Deferring communication for 1 days 0 hr 0 min 0 sec
2007-07-03 22:09:16 [climateprediction.net] Reason: requested by project

What happen?
Can I solve this problem?

---

There is no 64-bit CPDN client. Your XP-64 installation must be missing the 32-bit libraries. (I have XP-64 running on three of my boxes without issues.) Beyond that, I can\'t help.

5.10.7 has been \"recalled\". Bugs. Official version is, once again, 5.8.16. [Edit: Had 5.10.7 on two machines for awhile, until one had the \"lost icons\" crash. No difficulties communicating with the servers on either machine.]

---

"We have met the enemy and he is us." -- Pogo
Greetings from coastal Washington state, the scenic US Pacific Northwest.

This project doesn\'t have 64 bit applications, and the server software hasn\'t been updated to include the part that sends 32 bit apps to 64 bit systems.

Also, 5.10.7 is seriously faulty, and has been withdrwn from use.
It can delete all of the icons from the desktop, although rebooting will restore them.

edit
\'Pipped at the post\'.
:)

This project doesn\'t have 64 bit applications, and the server software hasn\'t been updated to include the part that sends 32 bit apps to 64 bit systems.

Also, 5.10.7 is seriously faulty, and has been withdrwn from use.
It can delete all of the icons from the desktop, although rebooting will restore them.

edit
\'Pipped at the post\'.
:)

Well I\'m with the same problem of \"platform not found\" and I use 5.10.8 version (Beta I think).

You say that the server hasn\'t been updated, so I ask if you know when it will be updated.

I can revert to the 32bit client but my experience with this 64bit version has been positive and now I have a must better memory management than I have with teh 32bit client on the same machine.

Best regards,

PAMF

---

The 5.10.8 has the same bug which causes your \'explorer\' to crash, and hence lose control of the PC. 5.10.13 might have resolved that, but it hasn\'t been confirmed yet.

I don\'t know when the serverside code will be updated.

---

I'm a volunteer and my views are my own.
News and Announcements and FAQ

The server software may get upgraded sometime in the next 2-3 weeks. It does take time, and the 2 staff people have a REALLY big work load.

The server software may get upgraded sometime in the next 2-3 weeks. It does take time, and the 2 staff people have a REALLY big work load.

I am getting the same error, and I can\'t report any work thusfar. Is there any word on upgrading the server? Or are there any other solutions?

Kind regards

Kenneth

---

There are some more urgent issues which are taking everyone\'s attention (code 22s and performance of the 5.44 client), once the developers are able to stop firefighting and have enough time to look at other issues then it\'ll get sorted.

I\'m running a 64-bit box without problems (openSUSE 10.2 64-bit) by using the 32-bit linux manager+client. Obviously 32-bit support needs to be enabled on the operating system. I gather the same solution will also work on Windows 64-bit.

---

I'm a volunteer and my views are my own.
News and Announcements and FAQ

I\'m running a 64-bit box without problems (openSUSE 10.2 64-bit) by using the 32-bit linux manager+client. Obviously 32-bit support needs to be enabled on the operating system. I gather the same solution will also work on Windows 64-bit.

I am running version 5.10.13, but that doesn\'t seem to be the latest version that can be downloaded from the CPDN website. Is it possible to revert to the 5.8.16 and keep the work units I am crunching?


EDIT
----
Seems it can\'t be done, I am getting this error:

Platform changed from windows_x86 to windows_x86_64 - resetting projects

What now?

---

I\'m using 5.10.8 32-bit Boinc manager on my 64-bit linux box, and that works fine. The nearest equivalent for your XP 64-bit system would be 5.10.13 32-bit (see the other posts in this thread). (http://boinc.berkeley.edu/download_all.php)

---

I'm a volunteer and my views are my own.
News and Announcements and FAQ

I\'m using 5.10.8 32-bit Boinc manager on my 64-bit linux box, and that works fine. The nearest equivalent for your XP 64-bit system would be 5.10.13 32-bit (see the other posts in this thread). (http://boinc.berkeley.edu/download_all.php)

I know that the 32-bit manager works on x64 (just tried it out) but since I used the 64-bit client, there seems to be no way to change back to 32-bit without loss of the WU\'s your calculating right now (see message above) and that bothers me. I don\'t want to throw these WU\'s in the recycle bin, since they matter to CPDN..

---

Are you getting this message when going from 5.10 64-bit to 5.8 32-bit, or when you\'re going from 5.10 64-bit to 5.10 32-bit?

---

I'm a volunteer and my views are my own.
News and Announcements and FAQ

Are you getting this message when going from 5.10 64-bit to 5.8 32-bit, or when you\'re going from 5.10 64-bit to 5.10 32-bit?

I am getting this error when I\'m going from 5.10 64-bit to 5.8 32-bit..
Thanks for looking into this matter.

---

If you go 5.10 64-bit to 5.10 32-bit that should sort it out (note that I haven\'t tried it myself, but I saw it mentioned in Boinc/Trac recently)

---

I'm a volunteer and my views are my own.
News and Announcements and FAQ

If you go 5.10 64-bit to 5.10 32-bit that should sort it out (note that I haven\'t tried it myself, but I saw it mentioned in Boinc/Trac recently)

MikeMarsUK, thanks for your advice. The project hasn\'t reset itself so that is a good thing. Right now the only remaining thing that bothers me is that I didn\'t seem to get any credit for the WU I uploaded and another one that is still in progress. (Normally I get 500 points a day, but right now I received zero points while I was unable to upload WU\'s for 4 days)

Do you know what causes this?

---

Once the trickle has been uploaded to the servers, it needs to wait for the nightly trickle + credit processing batch job to run. If you wait 24 hours you\'ll see it all appear (if you look against your model\'s result page, you\'ll see the trickles are already listed).

---

I'm a volunteer and my views are my own.
News and Announcements and FAQ

How / where in the client (5.10.13) can one see if a trickle up is ready to be send and how many? There\'s not a trace in the message logs. FTM i check BOINCView where the project tab says one is ready, but really it would be nice if it created an actual log entry in the client.

There ARE messages in the log!
They look like this:

2006-12-17 01:23:59 [climateprediction.net] Sending scheduler request to http://climateapps2.oucs.ox.ac.uk/cpdnboinc_cgi/cgi
2006-12-17 01:23:59 [climateprediction.net] Reason: To send trickle-up message
2006-12-17 01:23:59 [climateprediction.net] Note: not requesting new work or reporting results
2006-12-17 01:24:19 [climateprediction.net] Scheduler request to http://climateapps2.oucs.ox.ac.uk/cpdnboinc_cgi/cgi succeeded

And, for the record, the messages are archived in the BOINC folder as follows:
stdoutdae.txt contains ALL of the messages, and
stderrdae.txt contains only the error messages.

******************

Trickle_up files appear in the climateprediction.net folder, on about 3rd December of each year.
They do NOT appear in the Transfers tab, and are uploaded in seconds, so you may not see them appear in the folder unless you\'re watching at the exact right time, or you keep the Network option turned off.

If you\'re watching the Show graphics globe at the right time, you can also see a message about the yearly data being collected.

---

Backups: Here

Once the trickle has been uploaded to the servers, it needs to wait for the nightly trickle + credit processing batch job to run. If you wait 24 hours you\'ll see it all appear (if you look against your model\'s result page, you\'ll see the trickles are already listed).

MikeMarsUK, thanks for your advice on how to solve this issue! Everything is just fine right now!

Issue has been resolved very quickly and adequatly, congratulation to CPDN and MikeMarsUK especially.

---

There ARE messages in the log!
They look like this:

2006-12-17 01:23:59 [climateprediction.net] Sending scheduler request to http://climateapps2.oucs.ox.ac.uk/cpdnboinc_cgi/cgi
2006-12-17 01:23:59 [climateprediction.net] Reason: To send trickle-up message
2006-12-17 01:23:59 [climateprediction.net] Note: not requesting new work or reporting results
2006-12-17 01:24:19 [climateprediction.net] Scheduler request to http://climateapps2.oucs.ox.ac.uk/cpdnboinc_cgi/cgi succeeded

And, for the record, the messages are archived in the BOINC folder as follows:
stdoutdae.txt contains ALL of the messages, and
stderrdae.txt contains only the error messages.

******************

Trickle_up files appear in the climateprediction.net folder, on about 3rd December of each year.
They do NOT appear in the Transfers tab, and are uploaded in seconds, so you may not see them appear in the folder unless you\'re watching at the exact right time, or you keep the Network option turned off.

If you\'re watching the Show graphics globe at the right time, you can also see a message about the yearly data being collected.

Okay Les, thanks and u reminded me of posts on the BOINC developers forum where certain message stored in the stdoutdae.txt file but not all showing in the Message Tab. Never seen the trickle entries there under any version of BOINC 5.4, 5.8 and now 5.10.13, English language instance, but indeed they are in the disk stored archive.... cumbersome. At several times when crunching longer off line I did see a Zip file appear in the Transfer tab of about 5mb. Is there a rule on the number of trickles/years completed for that to happen (these are surely the ones needed and not the credit trickles)?

Separate question.... why can I post in this Q&A forum but not the big one (why are there 2 anyway?). Send mails, but like what I hear from others, the responsiveness to support mails is none or rather negatively toned. The question i had included was also the ambiguous postings of news inferring that parts of models are being send out again due manco.... Well I\'ve only reached 1985 starting Dec.06 with this WU, so is this a waste and should i ditch - ?

Now the scary bit.... read of models needing 5gb RAM/VM in future releases.... is that serious?

Feel free to split up this post and put them in separate topics.... seems this thread is all over the place.

cheers

Added: Result ID is 5997428 and Work Unit ID is 5737343.