CPU/SMP 高獎勵得分 選購/設定/除錯...集中綜合討論串
- 主題發起人 Raziel
- 開始日期
可惜了, 不知道失敗的原因? kazama大給的連結, 後面有一位zodac網友有提到他是用VMware在跑的說.....
小弟的2600K今天剛遞交了P6903, 得分257920, PPD超過70K, 就算沒接到P6903/P6904, 一般的bigadv wu的PPD也比在windows下高一點, 所以小弟專門folding的機器都會改成ubuntu...
OverFold
高級會員
- 已加入
- 7/29/09
- 訊息
- 541
- 互動分數
- 20
- 點數
- 18
唉~ 小弟再一次以VM試行Ubuntu於2500K 平台模擬12 Core取得6900跑BigAdv，
再一次遭逢滑鐵盧，又是到100%後表示Connect Server不能云云.....
[17:36:02] Project: 2685 (Run 2, Clo[17:36:02] Project: 2685 (Run 2, Clone 23, Gen 108)
[17:36:02]
[17:36:02] Entering M.D.
[17:36:08] Using Gromacs checkpoints
[17:36:09] Mapping NT from 12 to 12
[17:36:15] Resuming from checkpoint
[17:36:16] Verified work/wudata_01.log
[17:36:16] Verified work/wudata_01.trr
[17:36:16] Verified work/wudata_01.xtc
[17:36:16] Verified work/wudata_01.edr
[17:36:23] Completed 248483 out of 250000 steps (99%)
[18:02:12] Completed 250000 out of 250000 steps (100%)
[18:02:44] DynamicWrapper: Finished Work Unit: sleep=10000
[18:02:54] Finished Work Unit:
[18:02:54] - Reading up to 56507088 from "work/wudata_01.trr": Read 56507088
[18:02:59] trr file hash check passed.
[18:02:59] - Reading up to 45863088 from "work/wudata_01.xtc": Read 45863088
[18:03:04] xtc file hash check passed.
[18:03:04] edr file hash check passed.
[18:03:04] logfile size: 211496
[18:03:04] Leaving Run
[18:03:09] - Writing 102752020 bytes of core data to disk...
[18:03:19] ... Done.
[18:03:34] - Shutting down core
[18:03:34] Folding@home Core Shutdown: FINISHED_UNIT
[18:03:35] CoreStatus = 64 (100)
[18:03:35] Sending work to server
[18:03:35] Project: 2685 (Run 2, Clone 23, Gen 108)
[18:03:35] + Attempting to send results [July 27 18:03:35 UTC]
[18:03:36] - Couldn't send HTTP request to server ***[WHY?]***
[18:03:36] + Could not connect to Work Server (results)
[18:03:36] (171.67.108.22:8080)
[18:03:36] + Retrying using alternative port
[18:03:36] - Couldn't send HTTP request to server ***[WHY?WHY?]***
[18:03:36] + Could not connect to Work Server (results)
[18:03:36] (171.67.108.22:80)
[18:03:36] - Error: Could not transmit unit 01 (completed July 27) to work server.
[18:03:36] Keeping unit 01 in queue.
[18:03:36] Project: 2685 (Run 2, Clone 23, Gen 108)
[18:03:36] Could not transmit unit 01 to Collection server; keeping in queue.
[18:03:36] - Preparing to get new work unit...
[18:03:36] Cleaning up work directory ***[把我完成的WU delete了~]***
[18:03:36] + Attempting to get work packet
[18:03:36] Passkey found
[18:03:36] - Connecting to assignment server
[18:03:37] - Successful: assigned to (130.237.232.141).
[18:03:37] + News From Folding@Home: Welcome to Folding@Home
[18:03:38] Loaded queue successfully.
[18:05:37] Project: 2685 (Run 2, Clone 23, Gen 108)
我在99%進度時有將整個fah folder backup，未知抄回原位再從99%跑多次
未知是否有效呢？目前最擔心的是明天準備再提交跑了6903的2600K平台會
出現以上相同的問題我就真的損失慘重了。
再一次遭逢滑鐵盧，又是到100%後表示Connect Server不能云云.....
[17:36:02] Project: 2685 (Run 2, Clo[17:36:02] Project: 2685 (Run 2, Clone 23, Gen 108)
[17:36:02]
[17:36:02] Entering M.D.
[17:36:08] Using Gromacs checkpoints
[17:36:09] Mapping NT from 12 to 12
[17:36:15] Resuming from checkpoint
[17:36:16] Verified work/wudata_01.log
[17:36:16] Verified work/wudata_01.trr
[17:36:16] Verified work/wudata_01.xtc
[17:36:16] Verified work/wudata_01.edr
[17:36:23] Completed 248483 out of 250000 steps (99%)
[18:02:12] Completed 250000 out of 250000 steps (100%)
[18:02:44] DynamicWrapper: Finished Work Unit: sleep=10000
[18:02:54] Finished Work Unit:
[18:02:54] - Reading up to 56507088 from "work/wudata_01.trr": Read 56507088
[18:02:59] trr file hash check passed.
[18:02:59] - Reading up to 45863088 from "work/wudata_01.xtc": Read 45863088
[18:03:04] xtc file hash check passed.
[18:03:04] edr file hash check passed.
[18:03:04] logfile size: 211496
[18:03:04] Leaving Run
[18:03:09] - Writing 102752020 bytes of core data to disk...
[18:03:19] ... Done.
[18:03:34] - Shutting down core
[18:03:34] Folding@home Core Shutdown: FINISHED_UNIT
[18:03:35] CoreStatus = 64 (100)
[18:03:35] Sending work to server
[18:03:35] Project: 2685 (Run 2, Clone 23, Gen 108)
[18:03:35] + Attempting to send results [July 27 18:03:35 UTC]
[18:03:36] - Couldn't send HTTP request to server ***[WHY?]***
[18:03:36] + Could not connect to Work Server (results)
[18:03:36] (171.67.108.22:8080)
[18:03:36] + Retrying using alternative port
[18:03:36] - Couldn't send HTTP request to server ***[WHY?WHY?]***
[18:03:36] + Could not connect to Work Server (results)
[18:03:36] (171.67.108.22:80)
[18:03:36] - Error: Could not transmit unit 01 (completed July 27) to work server.
[18:03:36] Keeping unit 01 in queue.
[18:03:36] Project: 2685 (Run 2, Clone 23, Gen 108)
[18:03:36] Could not transmit unit 01 to Collection server; keeping in queue.
[18:03:36] - Preparing to get new work unit...
[18:03:36] Cleaning up work directory ***[把我完成的WU delete了~]***
[18:03:36] + Attempting to get work packet
[18:03:36] Passkey found
[18:03:36] - Connecting to assignment server
[18:03:37] - Successful: assigned to (130.237.232.141).
[18:03:37] + News From Folding@Home: Welcome to Folding@Home
[18:03:38] Loaded queue successfully.
[18:05:37] Project: 2685 (Run 2, Clone 23, Gen 108)
我在99%進度時有將整個fah folder backup，未知抄回原位再從99%跑多次
未知是否有效呢？目前最擔心的是明天準備再提交跑了6903的2600K平台會
出現以上相同的問題我就真的損失慘重了。
OverFold
高級會員
- 已加入
- 7/29/09
- 訊息
- 541
- 互動分數
- 20
- 點數
- 18
問題似乎解決了！WU成功Sent了！但腦海卻飄浮著數個問號???
overclock.net那篇教學中有一項題到，在建立Terminal Profile
"langouste"時，"Title and Command"內應鍵入如此：
/home/overfold/fah/langouste/langouste3 -l 8880
之後的之後再之後之後，首次執行fah6smp.sh要建立config.cfg時
文中刻意題到，Use proxy 要 say YES
Enter 127.0.0.1 for proxy Name
Enter 8880 for port
Say no for username and password.
Port 8880此項貌似要呼應langouste在localhost執行時需改以
此Port運作而非Default的Port 8080，教學既然如此相信也必定
有他的道理，亦從來無懷疑過會影響正常運作，直至見到那Forum
內"darksun20"亦反映出現WU完成後can't connect to server
，我才發現原來無論Config.cfg內Proxy設定是否行Port 8880，
實際上也能成功模擬12 Core並達到抓取Project 6903的目標，
當我將Proxy設定改回8080後再從99%Backup的WU再Run一次
，不久之後那個完成100%的WU已能如常地Sent出，再Restart
Ubuntu一次測試是否能模擬12 Core亦輕鬆做到。
到底那個Proxy設定是否必需，還是只為作弄人而設的呢？我半個月
前花6天完成的6903就是因這小誤會而被迫放棄200,000多萬積分
，如果當時我發現到用這個方法便能修復你說多好呢？對不對？
overclock.net那篇教學中有一項題到，在建立Terminal Profile
"langouste"時，"Title and Command"內應鍵入如此：
/home/overfold/fah/langouste/langouste3 -l 8880
之後的之後再之後之後，首次執行fah6smp.sh要建立config.cfg時
文中刻意題到，Use proxy 要 say YES
Enter 127.0.0.1 for proxy Name
Enter 8880 for port
Say no for username and password.
Port 8880此項貌似要呼應langouste在localhost執行時需改以
此Port運作而非Default的Port 8080，教學既然如此相信也必定
有他的道理，亦從來無懷疑過會影響正常運作，直至見到那Forum
內"darksun20"亦反映出現WU完成後can't connect to server
，我才發現原來無論Config.cfg內Proxy設定是否行Port 8880，
實際上也能成功模擬12 Core並達到抓取Project 6903的目標，
當我將Proxy設定改回8080後再從99%Backup的WU再Run一次
，不久之後那個完成100%的WU已能如常地Sent出，再Restart
Ubuntu一次測試是否能模擬12 Core亦輕鬆做到。
到底那個Proxy設定是否必需，還是只為作弄人而設的呢？我半個月
前花6天完成的6903就是因這小誤會而被迫放棄200,000多萬積分
，如果當時我發現到用這個方法便能修復你說多好呢？對不對？
我是按照guide裡面的設定proxy 8880 port, 有時會出現
Could not connect to Work Server (results),
不過讓程式多試幾次後還是可以將完成的wu遞交...
另外, 如果花了6天才完成P6903, 那應該就沒有bonus了, 前兩天看到7月20日的 Announcement
http://foldingforum.org/viewtopic.php?f=24&t=19180
"........The new preferred deadlines are 5 days for 6903, 5.6 for 6904. Final deadlines and k-factors remain unchanged."
所以2500K跑P6903/P6904可能太勉強了...
