Can't Merge Computers
Message boards : Macintosh : Can't Merge Computers
Author | Message | |
---|---|---|
Hi!
|
||
ID: 2334 | Rating: 0 | rate: / | ||
You can do it yourself by going to the host overview page on your account. There should be a merge button somewhere...
Hi! ____________ D@H the greatest project in the world... a while from now! |
||
ID: 2335 | Rating: 0 | rate: / | ||
You can do it yourself by going to the host overview page on your account. There should be a merge button somewhere... Unfortunately, BOINC changed my computer's OS name so that they don't match. This prevents me from merging the computers myself. ____________ John MacPro 2 x 2.66GHz Dual-Core Xeon | 2GB RAM | ATI x1900 | BOINC 5.9.5 |
||
ID: 2336 | Rating: 0 | rate: / | ||
In that case, there's not much hope. Boinc does not supply tools for server-side merging (yet?) and I don't want to mess with the database and break things... Maybe somebody can write a tool or script to do this.
You can do it yourself by going to the host overview page on your account. There should be a merge button somewhere... ____________ D@H the greatest project in the world... a while from now! |
||
ID: 2337 | Rating: 0 | rate: / | ||
Hi! The OS is not reported differently. They both show as "Darwin 8.8.1" And BOINC does not change your computer name. In any case, you can change your computer name back by going to System Preferences => Sharing. I think the issue is that you changed to a newer BOINC client at the same time your system name changed. The standard (5.4.*) BOINC client reports the CPU type differently form the new 5.8.5+. So when your reported CPU type and your system name changed at the same time, it caused the project to assume it is a new machine. Try changing one or both back to the old name/client; then crunch a WU and return it; and then try to merge them again. ____________ Dublin, CA Team SETI.USA |
||
ID: 2351 | Rating: 0 | rate: / | ||
Sorry. I wasn't specific. The new client reports the CPUs differently. Old BOINC said, "i386 MacPro1,1." New BOINC calls my Xeons, "GenuineIntel Intel(R) Xeon(R) CPU 5150 @ 2.66GHz [x86 Family 6 Model 15 Stepping 6] [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 MNI CX16 TPR]." I believe it is this difference that disallows the merging. ____________ John MacPro 2 x 2.66GHz Dual-Core Xeon | 2GB RAM | ATI x1900 | BOINC 5.9.5 |
||
ID: 2354 | Rating: 0 | rate: / | ||
"GenuineIntel Are you saying that the above string identifies the cpu in your mac? Hmmm, that might break our HR rules (we filter on i386 for macs). Have to check this carefully.... Seems that the new boinc client is causing a lot of trouble for us :-( Thanks! Andre ____________ D@H the greatest project in the world... a while from now! |
||
ID: 2355 | Rating: 0 | rate: / | ||
I believe it is this difference that disallows the merging. Right. Going back to 5.4.9 will solve that. Try going back, then completing a WU, and then try to merge them. I am still concerned that your machine name changed too. You can fix that in sharing. Finally, when I upgraded to 5.8.6, I did not have this problem. Yes, the description of my processor changed, but it did not create a new host entry in the host list on D@H. So I don't think the change in processor description *alone* caused the split for you. Perhaps it was the simultaneous change of host name and processor description. ____________ Dublin, CA Team SETI.USA |
||
ID: 2356 | Rating: 0 | rate: / | ||
Are you saying that the above string identifies the cpu in your mac? Hmmm, that might break our HR rules (we filter on i386 for macs). Have to check this carefully.... Yes. However, it has made no difference to my two macs. They are still getting, and successfully processing results. Feel free to look. ____________ Dublin, CA Team SETI.USA |
||
ID: 2357 | Rating: 0 | rate: / | ||
"GenuineIntel No problem. I would like to go back to 5.8.2, but it isn't posted anymore, so I'm running 5.8.8. 5.4.9 doesn't utilize the new security features found in 5.5.4 and later. Also, I had issues running 5.4.9 earlier, though that may have been due to other things. Since I started using 5.8.2, I've had zero bad workunits. Not sure how 5.8.8 is doing yet... :-) Let me know by reposting here if I should attempt to install 5.4.9. I'll run 5.8.8 until then. ____________ John MacPro 2 x 2.66GHz Dual-Core Xeon | 2GB RAM | ATI x1900 | BOINC 5.9.5 |
||
ID: 2358 | Rating: 0 | rate: / | ||
I believe it is this difference that disallows the merging. I changed my mac name back a few days ago, so it is now listed as the same. ____________ John MacPro 2 x 2.66GHz Dual-Core Xeon | 2GB RAM | ATI x1900 | BOINC 5.9.5 |
||
ID: 2359 | Rating: 0 | rate: / | ||
"GenuineIntel Oops, and I thought, this was done on your request maybe after the discussion about the p_capabilities here in the forum. Starting with version 5.8.5 the p_capabilities tag disappeared in the client_state.xml and the processor identification by BOINC was extended by one tag showing the Family/Model/Stepping (= cpuid) and one tag listing the processor features. The features list is far more verbose on Linux than on Windows. And it seems on MacOSX it is equally verbose (no surprise, after all it is a *nix in it's core). According to this post over at the NanoHive forum, the list of processor features is even dependent on the version of the OS one is using. For example, Windows XP does not report SSE3 apparently while Windows Vista does report it. I don't know if this new BOINC 'feature' will make it into the recommended stable version, but if yes, I would suggest to include also the Vendor (Genuine Intel, AuthenticAMD,...) into every processor identification by BOINC. Because, that way you can really identify each cpu without any doubt. Vendor and cpuid combined are 100% conclusive. But the BOINC devs should make sure, that it will be independent from the OS version. The latter might not be very easy, though. Btw, here's a victim of the new BOINC 'feature' on the Windows platform: Workunit ID 20694 This workunit was shared by two Pentium III and one Core 2. The Core 2 was using BOINC 5.8.5 and received no credit for its result. And speaking of unusual quorums: I excavated my old 166 MHz Pentium MMX (which is a 586 Pentium) and it was paired with a Pentium M and a Pentium 4 Xeon: Workunit ID 24659 Not that it matters much (it's way too old for serious crunching) but it shows how difficult it can be to maintain and adjust the HR rules for all those different x86 breeds out there. Regards Alex ____________ |
||
ID: 2360 | Rating: 0 | rate: / | ||
I would like to go back to 5.8.2, but it isn't posted anymore, so I'm running 5.8.8. John, by this URL you can also access older versions of BOINC: http://boinc.berkeley.edu/dl/?C=M;O=D Is the processor identification back to the 'old' pre-5.8.5 style with 5.8.8? Regards Alex |
||
ID: 2361 | Rating: 0 | rate: / | ||
I would like to go back to 5.8.2, but it isn't posted anymore, so I'm running 5.8.8. I don't think so. I just reinstalled 5.4.9, and after a few minutes, checked my computers. It showed that they are back to the same, so I merged them. Don't know if just changing my client for a few minutes (with no work available) did that, or if someone here at Docking manually changed it... Maybe I should try going back to 5.8.8? Is there any speed/quality difference between 5.4.9 and 5.8.8? ____________ John MacPro 2 x 2.66GHz Dual-Core Xeon | 2GB RAM | ATI x1900 | BOINC 5.9.5 |
||
ID: 2362 | Rating: 0 | rate: / | ||
I would like to go back to 5.8.2, but it isn't posted anymore, so I'm running 5.8.8. Oh, thanks for the link! I'll try 5.8.4 again. ____________ John MacPro 2 x 2.66GHz Dual-Core Xeon | 2GB RAM | ATI x1900 | BOINC 5.9.5 |
||
ID: 2363 | Rating: 0 | rate: / | ||
Is there any speed/quality difference between 5.4.9 and 5.8.8? Well, the Linux 5.8.x client has improved benchmark values (over 5.4.x), which brings the credits closer to computers running Windows. Of course only on projects, which use the benchmarks and the claimed credits (Docking still does it that way, but fixed credits have been announced already to be introduced later). Other than that, I can't see any 'must-have' features in the 5.8.x until this becomes the stable version of BOINC. I have switched to 5.8.x because of some problems with DepSpid. This project is running on a server which is getting its IP by DynDNS, so the IP is changing every day and the new libcurl in BOINC 5.8.5+ can cope a little bit better with it (but it's still far from perfect). Regards Alex |
||
ID: 2364 | Rating: 0 | rate: / | ||
...I have switched to 5.8.x because of some problems with DepSpid. This project is running on a server which is getting its IP by DynDNS, so the IP is changing every day and the new libcurl in BOINC 5.8.5+ can cope a little bit better with it (but it's still far from perfect). Thanks for the info! I'll try 5.8.4 for awhile and see if I messed anything up. Of course, now Docking is out of work, so I guess I'll try the others for a few days. No biggie. I think I still have around 3000 units pending. If I could just get past Memo's computers! I'm within a few thousand units! WooHoo! This is kinda fun! I don't feel lost in the multitudes here. ____________ John MacPro 2 x 2.66GHz Dual-Core Xeon | 2GB RAM | ATI x1900 | BOINC 5.9.5 |
||
ID: 2365 | Rating: 0 | rate: / | ||
I don't think so. I just reinstalled 5.4.9, and after a few minutes, checked my computers. It showed that they are back to the same, so I merged them. Don't know if just changing my client for a few minutes (with no work available) did that, or if someone here at Docking manually changed it... Great! Not that they are merged again, try going back to 5.8.8. My guess is that you will not end up with duplicate hosts this time. And even if it does, you now know how to fix it. ____________ Dublin, CA Team SETI.USA |
||
ID: 2366 | Rating: 0 | rate: / | ||
OK. Now I'm running 5.8.8, and my one computer has the long CPU name. Hopefully, someone will let me know if that's a problem for Docking? PS... Thanks for the help, everyone! ____________ John MacPro 2 x 2.66GHz Dual-Core Xeon | 2GB RAM | ATI x1900 | BOINC 5.9.5 |
||
ID: 2367 | Rating: 0 | rate: / | ||
I don't think so. I just reinstalled 5.4.9, and after a few minutes, checked my computers. It showed that they are back to the same, so I merged them. Don't know if just changing my client for a few minutes (with no work available) did that, or if someone here at Docking manually changed it... Not us :-)
There shouldn't be any difference in app behavior; the only difference might be benchmarking results. Andre ____________ D@H the greatest project in the world... a while from now! |
||
ID: 2374 | Rating: 0 | rate: / | ||
Yep, this new functionality is definitely breaking our HR rules. This will be hard to solve until everybody goes to the new format (maybe we will have to enforce a min. boinc client version soon). Family x before was always a PII/III, now it can be everything...
Btw, here's a victim of the new BOINC 'feature' on the Windows platform: ____________ D@H the greatest project in the world... a while from now! |
||
ID: 2375 | Rating: 0 | rate: / | ||
The new client has been officially released. Message from Rom:
|
||
ID: 2376 | Rating: 0 | rate: / | ||
The new client has been officially released. Message from Rom: Thanks, Andre! I'll just switch to that and leave it for awhile! <edit> I guess the release version is 5.8.8, not 5.8; so it's what I'm already running. I reinstalled it anyway, just in case! ____________ John MacPro 2 x 2.66GHz Dual-Core Xeon | 2GB RAM | ATI x1900 | BOINC 5.9.5 |
||
ID: 2378 | Rating: 0 | rate: / | ||
Great. 5.8.x on Mac and Windows should work fine (I hope). We'll keep on monitoring the system...
The new client has been officially released. Message from Rom: ____________ D@H the greatest project in the world... a while from now! |
||
ID: 2379 | Rating: 0 | rate: / | ||
Message boards : Macintosh : Can't Merge Computers
Database Error: The MySQL server is running with the --read-only option so it cannot execute this statement
array(3) { [0]=> array(7) { ["file"]=> string(47) "/boinc/projects/docking/html_v2/inc/db_conn.inc" ["line"]=> int(97) ["function"]=> string(8) "do_query" ["class"]=> string(6) "DbConn" ["object"]=> object(DbConn)#29 (2) { ["db_conn"]=> resource(72) of type (mysql link persistent) ["db_name"]=> string(7) "docking" } ["type"]=> string(2) "->" ["args"]=> array(1) { [0]=> &string(51) "update DBNAME.thread set views=views+1 where id=166" } } [1]=> array(7) { ["file"]=> string(48) "/boinc/projects/docking/html_v2/inc/forum_db.inc" ["line"]=> int(60) ["function"]=> string(6) "update" ["class"]=> string(6) "DbConn" ["object"]=> object(DbConn)#29 (2) { ["db_conn"]=> resource(72) of type (mysql link persistent) ["db_name"]=> string(7) "docking" } ["type"]=> string(2) "->" ["args"]=> array(3) { [0]=> object(BoincThread)#3 (16) { ["id"]=> string(3) "166" ["forum"]=> string(1) "7" ["owner"]=> string(3) "188" ["status"]=> string(1) "0" ["title"]=> string(21) "Can't Merge Computers" ["timestamp"]=> string(10) "1170369234" ["views"]=> string(4) "1205" ["replies"]=> string(2) "23" ["activity"]=> string(23) "3.2436020022893997e-124" ["sufferers"]=> string(1) "0" ["score"]=> string(1) "0" ["votes"]=> string(1) "0" ["create_time"]=> string(10) "1170175465" ["hidden"]=> string(1) "0" ["sticky"]=> string(1) "0" ["locked"]=> string(1) "0" } [1]=> &string(6) "thread" [2]=> &string(13) "views=views+1" } } [2]=> array(7) { ["file"]=> string(63) "/boinc/projects/docking/html_v2/user/community/forum/thread.php" ["line"]=> int(184) ["function"]=> string(6) "update" ["class"]=> string(11) "BoincThread" ["object"]=> object(BoincThread)#3 (16) { ["id"]=> string(3) "166" ["forum"]=> string(1) "7" ["owner"]=> string(3) "188" ["status"]=> string(1) "0" ["title"]=> string(21) "Can't Merge Computers" ["timestamp"]=> string(10) "1170369234" ["views"]=> string(4) "1205" ["replies"]=> string(2) "23" ["activity"]=> string(23) "3.2436020022893997e-124" ["sufferers"]=> string(1) "0" ["score"]=> string(1) "0" ["votes"]=> string(1) "0" ["create_time"]=> string(10) "1170175465" ["hidden"]=> string(1) "0" ["sticky"]=> string(1) "0" ["locked"]=> string(1) "0" } ["type"]=> string(2) "->" ["args"]=> array(1) { [0]=> &string(13) "views=views+1" } } }query: update docking.thread set views=views+1 where id=166