Hi;
Whenever either of my iMacs (both Intel Core 2 Machines), I get the following error message:
"Message from server: platform'x86_64-apple-darwin' not found

What can I do to resolve this?

Thanks;
Larry

---

LarryB56

"Message from server: platform'x86_64-apple-darwin' not found

What can I do to resolve this?

Not much you can do but wait - although reports of it are sprinkled through out several of the "post crash" threads I don't think that there are enough of us OSX users around to have even made the admin's radar yet.

Are people still having this issue? We never had an x86_64-apple-darwin platform but the boinc client should get the 32-bit version, i686-apple-darwin.

Are people still having this issue? We never had an x86_64-apple-darwin platform but the boinc client should get the 32-bit version, i686-apple-darwin.

I will confirm that as of 1330 CST I am still getting that message.

I can not be absolutely certain about it but I think that the above message is actually more related to the BOINC client than your specific MiniRosetta app.

I base this on the fact that the fist time I see the X86_64 is in the startup message (and for some reason I am having NO luck copying out of the message window)

Is goes like:

Starting BOINC Client 6.10.58 for x86_64-apple-darwin platform

so when you get this message, your boinc client doesn't use the 32bit mini application version instead? it should.

Same issue.
I have a Xserve running 10.5.8 OSX and a Power Mac Intel running 10.85.8 and a Minnie running 10.6 and all 3 of these are saying the following "platform'x86_64-apple-darwin' not found"

Fri Jan 14 18:00:23 2011 rosetta@home Message from server: platform 'x86_64-apple-darwin' not found


This is the error that I am seeing.....

"Message from server: platform'x86_64-apple-darwin' not found

What can I do to resolve this?

Not much you can do but wait - although reports of it are sprinkled through out several of the "post crash" threads I don't think that there are enough of us OSX users around to have even made the admin's radar yet.

I also echo the message, my good mac cannot do anything, same with my ubuntu...

Are people still having this issue? We never had an x86_64-apple-darwin platform but the boinc client should get the 32-bit version, i686-apple-darwin.

I never paid attention when I at first started to utilize my Mac but ever since the crash on Dec'28th, this keeps coming up, if the machine is/was able to reach a server at all...
And this has been mentioned by at least one user every day since... :-(

Last contact was on Jan'5th apprently, this is what the host listing on your end says

1394632 TPCBFs-iMac.local 302.31 11,211 GenuineIntel
Intel(R) Core(TM)2 Extreme CPU X7900 @ 2.80GHz [x86 Family 6 Model 15 Stepping 10] Darwin
10.5.0 52 5 Jan 2011 7:35:02 UTC

Ralf

David E K asked:

so when you get this message, your boinc client doesn't use the 32bit mini application version instead? it should.

I have been digging through my MAC system tonight and I can confirm that the actual application executables are 32 bit apps.

I have spent several hours doing a quick survey of the BOINC server and client source code. What I have found is that when the client contacts the server the client will identify what type of PLATFORM it is.

Please note that what is being identified is the PLATFORM not the client software. (hey its about 3:30 in the morning here and I apologize if my writing is a bit blunt - I'm trying not to confuse myself)

When it starts up the client will grab the version of the OS it is running on - 10.6.x will parse out to x86_64-apple-darwin

It appears that when the SERVER is started he will build a table of supported platforms using the the routine contained in "get_platforms.inc" - I'm a kernel / driver kind of guy and my PHP is weak to say the least but it looks like he pulls the information to build the table from the directory:

/home/boincadm/boinc/doc/platforms/

on the server. Do you have such a directory on your server and if so, could there be a missing file or entry?

I also see references to a list of supported platforms contained in an XML document get_project_config.php but I am not sure if this code path is ever used past the point f account creation.

In any case, what it looks like is that the CLIENT shoots a message to the SERVER saying in effect - "incoming from a "you fill in the OS" platform.

If the SERVER can not find that PLATFORM type in his table he shuts things down with the platform not supported message.

I'm not sure if any of this will be useful in putting a finger on the issue at hand - I may be completely out in left field - after all its late and there is a metric truckload of code and scripts here.

CH

so when you get this message, your boinc client doesn't use the 32bit mini application version instead? it should.

Please note that it is not merely that we can't get new tasks; the server is not accepting our reports of workunits completed during the outage either.


Best,
Snags

same with me here:


Sa 15 Jan 12:25:20 2011 rosetta@home update requested by user
Sa 15 Jan 12:25:21 2011 rosetta@home Sending scheduler request: Requested by user.
Sa 15 Jan 12:25:21 2011 rosetta@home Reporting 4 completed tasks, requesting new tasks for CPU and GPU
Sa 15 Jan 12:25:24 2011 rosetta@home Scheduler request completed: got 0 new tasks
Sa 15 Jan 12:25:24 2011 rosetta@home Message from server: platform 'x86_64-apple-darwin' not found


the tasks that are to be reported also ran out today...
and it took them a couple of days to be uploaded, with no obvious reason for fail/success.

greez

Same with me:

Sat 15 Jan 08:25:39 2011 Processor: 8 GenuineIntel Intel(R) Xeon(R) CPU W3530 @ 2.80GHz [x86 Family 6 Model 26 Stepping 5]
Sat 15 Jan 08:25:39 2011 Processor features: FPU VME DE PSE TSC MSR PAE MCE CX8 APIC SEP MTRR PGE MCA CMOV PAT PSE36 CLFSH DS ACPI MMX FXSR SSE SSE2 SS HTT TM SSE3 MON DSCPL VMX EST TM2 SSSE3 CX16 TPR PDCM SSE4.1 SSE4.2 POPCNT
Sat 15 Jan 08:25:39 2011 OS: Mac OS X 10.6.5 (Darwin 10.5.0)
Sat 15 Jan 08:25:39 2011 Memory: 7.00 GB physical, 501.63 GB virtual
Sat 15 Jan 08:25:39 2011 Disk: 931.19 GB total, 501.38 GB free
Sat 15 Jan 08:25:39 2011 Local time is UTC +8 hours
Sat 15 Jan 08:25:39 2011 No usable GPUs found
Sat 15 Jan 08:25:40 2011 rosetta@home URL https://boinc.bakerlab.org/rosetta/; Computer ID 1402444; resource share 100
Sat 15 Jan 08:25:40 2011 rosetta@home General prefs: from rosetta@home (last modified 14-Sep-2010 22:53:07)
Sat 15 Jan 08:25:40 2011 rosetta@home Computer location: home
Sat 15 Jan 08:25:40 2011 rosetta@home General prefs: no separate prefs for home; using your defaults
Sat 15 Jan 08:25:40 2011 Reading preferences override file
Sat 15 Jan 08:25:40 2011 Preferences:
Sat 15 Jan 08:25:40 2011 max memory usage when active: 3584.00MB
Sat 15 Jan 08:25:40 2011 max memory usage when idle: 6451.20MB
Sat 15 Jan 08:25:40 2011 max disk usage: 10.00GB
Sat 15 Jan 12:24:44 2011 rosetta@home Sending scheduler request: To fetch work.
Sat 15 Jan 12:24:44 2011 rosetta@home Requesting new tasks
Sat 15 Jan 12:24:47 2011 rosetta@home Scheduler request completed: got 0 new tasks
Sat 15 Jan 12:24:47 2011 rosetta@home Message from server: platform 'x86_64-apple-darwin' not found

---

Did your 64bit Macs work with R@h before the disk failure? The server has not changed so I'm a bit puzzled. The client should just get the 32bit version. I'll look into adding the 64bit version soon regardless.

Did your 64bit Macs work with R@h before the disk failure? The server has not changed so I'm a bit puzzled. The client should just get the 32bit version. I'll look into adding the 64bit version soon regardless.

Yup.

Did your 64bit Macs work with R@h before the disk failure? The server has not changed so I'm a bit puzzled. The client should just get the 32bit version. I'll look into adding the 64bit version soon regardless.

To be a little more specific let's take a look at one of my Mac systems - 1312275

This system has been up and running Snow Leopard since I purchased it about 6 months ago.

It has accumulated about 544K credits during that time.

Looking at it I can see that the last time it successfully contacted the server was 1/5/2011

One interesting thing is that the "My Computers" page on Rosetta shows this as a Darwin 10.5.0 system - while doing a "uname -a" on the system itself shows it running Darwin 10.6.0

No OS version upgrade has been done to any of my Macs recently.

Sat Jan 15 18:46:59 2011 Starting BOINC client version 6.10.58 for x86_64-apple-darwin
Sat Jan 15 18:46:59 2011 log flags: file_xfer, sched_ops, task
Sat Jan 15 18:46:59 2011 Libraries: libcurl/7.19.7 OpenSSL/0.9.7l zlib/1.2.3 c-ares/1.6.0
Sat Jan 15 18:46:59 2011 Data directory: /Library/Application Support/BOINC Data
Sat Jan 15 18:46:59 2011 Processor: 8 GenuineIntel Intel(R) Core(TM) i7 CPU 960 @ 3.20GHz [x86 Family 6 Model 26 Stepping 5]
Sat Jan 15 18:46:59 2011 Processor features: FPU VME DE PSE TSC MSR PAE MCE CX8 APIC SEP MTRR PGE MCA CMOV PAT PSE36 CLFSH DS ACPI MMX FXSR SSE SSE2 SS HTT TM SSE3 MON DSCPL VMX EST TM2 SSSE3 CX16 TPR PDCM SSE4.1 SSE4.2 POPCNT
Sat Jan 15 18:46:59 2011 OS: Mac OS X 10.6.4 (Darwin 10.4.0)
Sat Jan 15 18:46:59 2011 Memory: 12.00 GB physical, 88.15 GB virtual
Sat Jan 15 18:46:59 2011 Disk: 111.47 GB total, 87.90 GB free
Sat Jan 15 18:46:59 2011 Local time is UTC -6 hours
Sat Jan 15 18:46:59 2011 No usable GPUs found
Sat Jan 15 18:46:59 2011 No general preferences found - using BOINC defaults
Sat Jan 15 18:46:59 2011 Preferences:
Sat Jan 15 18:46:59 2011 max memory usage when active: 6144.00MB
Sat Jan 15 18:46:59 2011 max memory usage when idle: 11059.20MB
Sat Jan 15 18:46:59 2011 max disk usage: 10.00GB
Sat Jan 15 18:46:59 2011 don't use GPU while active
Sat Jan 15 18:46:59 2011 suspend work if non-BOINC CPU load exceeds 25 %
Sat Jan 15 18:46:59 2011 (to change preferences, visit the web site of an attached project, or select Preferences in the Manager)
Sat Jan 15 18:46:59 2011 Not using a proxy
Sat Jan 15 18:46:59 2011 This computer is not attached to any projects
Sat Jan 15 18:46:59 2011 Visit http://boinc.berkeley.edu for instructions
Sat Jan 15 18:47:12 2011 Fetching configuration file from https://boinc.bakerlab.org/rosetta/get_project_config.php
Sat Jan 15 18:47:27 2011 Running CPU benchmarks
Sat Jan 15 18:47:27 2011 Suspending computation - running CPU benchmarks
Sat Jan 15 18:47:58 2011 Benchmark results:
Sat Jan 15 18:47:58 2011 Number of CPUs: 8
Sat Jan 15 18:47:58 2011 3348 floating point MIPS (Whetstone) per CPU
Sat Jan 15 18:47:58 2011 5404 integer MIPS (Dhrystone) per CPU
Sat Jan 15 18:47:59 2011 Resuming computation
Sat Jan 15 18:48:01 2011 rosetta@home Master file download succeeded
Sat Jan 15 18:48:06 2011 rosetta@home Sending scheduler request: Project initialization.
Sat Jan 15 18:48:06 2011 rosetta@home Requesting new tasks
Sat Jan 15 18:48:09 2011 rosetta@home Scheduler request completed: got 0 new tasks
Sat Jan 15 18:48:09 2011 rosetta@home Message from server: platform 'x86_64-apple-darwin' not found

Did your 64bit Macs work with R@h before the disk failure? The server has not changed so I'm a bit puzzled. The client should just get the 32bit version. I'll look into adding the 64bit version soon regardless.

Yes, for the last two years. The last workunit received is from January 4th. The computer has been running on Leopard (currently 10.5.8) the whole time with the last update of any kind being a security update last November. BOINC hasn't been updated in quite a while either; I'm still on 6.10.21.


Best,
Snags

Did your 64bit Macs work with R@h before the disk failure? The server has not changed so I'm a bit puzzled. The client should just get the 32bit version. I'll look into adding the 64bit version soon regardless.

My setup worked well with R@h before the disk failure. Nothing has changed at my end since the last successfulk R@h task. As part of my initial trouble-shooting I reset the project, so can't quickly determine when the last R@h task was crunched here.

---

Okay, I think I fixed this. It was totally my fault and I'm sorry. After looking into this issue, I realized that the scheduler binaries (which I had to recompile after the crash) was not from the same source as before the crash. I updated the binaries and they should now recognize the "alt_platform". Thanks for all your input on this issue.