Difference between revisions of "Atlas:DataTransferTests"

Un article de lcgwiki.
Jump to: navigation, search
 
(Functional Test 5 Aug 2008)
 
(2 intermediate revisions by the same user not shown)
Ligne 1: Ligne 1:
* Functional Test 11 Jun 2008
+
= Functional Tests =
 +
== Functional Test 5 Aug 2008 ==
 +
There were many errors observed between 14h10 - 14h40.
 +
<pre>
 +
[SOURCE error during PREPARATION phase: [REQUEST_TIMEOUT] failed to prepare source file in 180 seconds] Source Host [ccsrm.in2p3.fr]
 +
</pre>
 +
 
 +
* A similar effect was observed also in CCRC08, but it was difficult to summarize at the time.
 +
 
 +
 
 +
<p>
 +
The files were transferred successfully in the end, but
 +
before the successfull transfers, there were many attempts
 +
with the error.
 +
 
 +
<br>
 +
<br>
 +
 
 +
From the graphs below;
 +
* the first 10 minutes: many transfers fail and few succeed
 +
* the next 20 minutes: still many transfers fail but success rate increases
 +
* the next 10 minutes: less failure rate with less transfers (most of the files have been transferred)
 +
 
 +
<pre>
 +
  Site                      14h00  14h10  14h20  14h30  14h40  14h50
 +
  ------------------------------------------------------------------------
 +
  BEIJING-LCG2_DATADISK      0/20    8/36  11/48  16/29    8/13    5/1
 +
  GRIF-LAL_DATADISK          1/18  17/27  18/29  18/11    2/2    1/0
 +
  GRIF-LAL_PRODDISK          0/3    2/9    2/10    3/8    1/2    1/0
 +
  GRIF-SACLAY_DATADISK      0/12    7/30  10/41  15/37  11/23    8/13
 +
  IN2P3-CPPM_DATADISK        0/12    4/16    5/20    6/9    2/5    1/1
 +
  IN2P3-CPPM_PRODDISK        0/6    0/7    1/5    1/3    1/0    0/0
 +
  IN2P3-LAPP_DATADISK        1/6    3/13    6/18    5/12    3/5    0/0
 +
  IN2P3-LPC_DATADISK        1/1    12/14  13/14  12/13    1/0    0/0
 +
  IN2P3-LPSC_DATADISK        0/6    2/8    2/6    2/2    0/0    0/0
 +
  RO-02-NIPNE_DATADISK      0/6    4/15    6/15    6/9    2/0    0/0
 +
  RO-07-NIPNE_DATADISK      0/4    0/4    0/4    0/0    0/0    0/0
 +
  ------------------------------------------------------------------------
 +
  total                      3/94  59/179  74/210  84/133  31/50  16/15
 +
</pre>
 +
 
 +
* 49 errors for BEIJING-LCG2_DATADISK [http://dashb-atlas-data.cern.ch/dashboard/request.py/file-placement?site=BEIJING-LCG2_DATADISK&fromDate=2008-08-05%2012:00:00&toDate=2008-08-05%2016:00:00&state=ATTEMPT_DONE&errorType=180]
 +
* 29 errors for GRIF-LAL_DATADISK [http://dashb-atlas-data.cern.ch/dashboard/request.py/file-placement?site=GRIF-LAL_DATADISK&fromDate=2008-08-05%2012:00:00&toDate=2008-08-05%2016:00:00&state=ATTEMPT_DONE&errorType=180]
 +
* 11 errors for  GRIF-LAL_PRODDISK [http://dashb-atlas-data.cern.ch/dashboard/request.py/file-placement?site=GRIF-LAL_PRODDISK&fromDate=2008-08-05%2012:00:00&toDate=2008-08-05%2016:00:00&state=ATTEMPT_DONE&errorType=180]
 +
* 56 errors for GRIF-SACLAY_DATADISK [http://dashb-atlas-data.cern.ch/dashboard/request.py/file-placement?site=GRIF-SACLAY_DATADISK&fromDate=2008-08-05%2012:00:00&toDate=2008-08-05%2016:00:00&state=ATTEMPT_DONE&errorType=180]
 +
* 21 errors for IN2P3-CPPM_DATADISK [http://dashb-atlas-data.cern.ch/dashboard/request.py/file-placement?site=IN2P3-CPPM_DATADISK&fromDate=2008-08-05%2012:00:00&toDate=2008-08-05%2016:00:00&state=ATTEMPT_DONE&errorType=180]
 +
* 15 errors for IN2P3-CPPM_PRODDISK [http://dashb-atlas-data.cern.ch/dashboard/request.py/file-placement?site=IN2P3-CPPM_PRODDISK&fromDate=2008-08-05%2012:00:00&toDate=2008-08-05%2016:00:00&state=ATTEMPT_DONE&errorType=180]
 +
* 18 errors for IN2P3-LAPP_DATADISK [http://dashb-atlas-data.cern.ch/dashboard/request.py/file-placement?site=IN2P3-LAPP_DATADISK&fromDate=2008-08-05%2012:00:00&toDate=2008-08-05%2016:00:00&state=ATTEMPT_DONE&errorType=180]
 +
* 14 errors for IN2P3-LPC_DATADISK  [http://dashb-atlas-data.cern.ch/dashboard/request.py/file-placement?site=IN2P3-LPC_DATADISK&fromDate=2008-08-05%2012:00:00&toDate=2008-08-05%2016:00:00&state=ATTEMPT_DONE&errorType=180]
 +
* 8 errors for IN2P3-LPSC_DATADISK [http://dashb-atlas-data.cern.ch/dashboard/request.py/file-placement?site=IN2P3-LPSC_DATADISK&fromDate=2008-08-05%2012:00:00&toDate=2008-08-05%2016:00:00&state=ATTEMPT_DONE&errorType=180]
 +
* 15 errors for RO-02-NIPNE_DATADISK [http://dashb-atlas-data.cern.ch/dashboard/request.py/file-placement?site=RO-02-NIPNE_DATADISK&fromDate=2008-08-05%2012:00:00&toDate=2008-08-05%2016:00:00&state=ATTEMPT_DONE&errorType=180]
 +
* 6 errors for RO-07-NIPNE_DATADISK [http://dashb-atlas-data.cern.ch/dashboard/request.py/file-placement?site=RO-07-NIPNE_DATADISK&fromDate=2008-08-05%2012:00:00&toDate=2008-08-05%2016:00:00&state=ATTEMPT_DONE&errorType=180]
 +
* RO-07-NIPNE was in downtime[https://goc.gridops.org/downtime/list?id=9155348], therefore zero success. But we still see this type of error in source preparation.
 +
* GRIF-LPNHE was in downtime [https://goc.gridops.org/downtime/list?id=9055364] and FTS channel was active, therefore many destination errors. (zero success rate)
 +
* TOKYO was in downtime [https://goc.gridops.org/downtime/list?id=9105364] and FTS channel was inactive, therefore no transfers.
 +
 
 +
* Total 242 errors of this type between 12h00 - 16h00, mostly between 14h00-15h00.
 +
 
 +
* there seem to be some inconsistency in the dashboard information. Thus the above numbers may not be exact.
 +
 
 +
 
 +
 
 +
; Throughput (MB/s) [http://dashb-atlas-data.cern.ch/dashboard/request.py/site?name=LYON&fromDate=2008-08-05%2012:00&toDate=2008-08-05%2016:00] : [[Image:LYON.throughput.14400.20080805.1600.png]]
 +
; Number of files  [http://dashb-atlas-data.cern.ch/dashboard/request.py/site?name=LYON&fromDate=2008-08-05%2012:00&toDate=2008-08-05%2016:00] : [[Image:LYON.num file xs.14400.20080805.1600.png]]
 +
; Number of transfer errors  [http://dashb-atlas-data.cern.ch/dashboard/request.py/site?name=LYON&fromDate=2008-08-05%2012:00&toDate=2008-08-05%2016:00] : [[Image:LYON.num file xs error.14400.20080805.1600.png]]
 +
 
 +
== Functional Test 11 Jun 2008 ==
 
** presentation at ADC operations http://indico.cern.ch/materialDisplay.py?contribId=5&materialId=slides&confId=36222
 
** presentation at ADC operations http://indico.cern.ch/materialDisplay.py?contribId=5&materialId=slides&confId=36222
 +
 +
= FDR =
 +
* [[Atlas:FDR2]]
 +
= CCRC08 =
 +
* [[Atlas:CCRC08May]]

Latest revision as of 17:53, 5 août 2008

Functional Tests

Functional Test 5 Aug 2008

There were many errors observed between 14h10 - 14h40.

[SOURCE error during PREPARATION phase: [REQUEST_TIMEOUT] failed to prepare source file in 180 seconds] Source Host [ccsrm.in2p3.fr]
  • A similar effect was observed also in CCRC08, but it was difficult to summarize at the time.


The files were transferred successfully in the end, but before the successfull transfers, there were many attempts with the error.

From the graphs below;

  • the first 10 minutes: many transfers fail and few succeed
  • the next 20 minutes: still many transfers fail but success rate increases
  • the next 10 minutes: less failure rate with less transfers (most of the files have been transferred)
  Site                      14h00   14h10   14h20   14h30   14h40   14h50
  ------------------------------------------------------------------------
  BEIJING-LCG2_DATADISK      0/20    8/36   11/48   16/29    8/13    5/1
  GRIF-LAL_DATADISK          1/18   17/27   18/29   18/11    2/2     1/0
  GRIF-LAL_PRODDISK          0/3     2/9     2/10    3/8     1/2     1/0
  GRIF-SACLAY_DATADISK       0/12    7/30   10/41   15/37   11/23    8/13
  IN2P3-CPPM_DATADISK        0/12    4/16    5/20    6/9     2/5     1/1
  IN2P3-CPPM_PRODDISK        0/6     0/7     1/5     1/3     1/0     0/0
  IN2P3-LAPP_DATADISK        1/6     3/13    6/18    5/12    3/5     0/0
  IN2P3-LPC_DATADISK         1/1    12/14   13/14   12/13    1/0     0/0
  IN2P3-LPSC_DATADISK        0/6     2/8     2/6     2/2     0/0     0/0
  RO-02-NIPNE_DATADISK       0/6     4/15    6/15    6/9     2/0     0/0
  RO-07-NIPNE_DATADISK       0/4     0/4     0/4     0/0     0/0     0/0
  ------------------------------------------------------------------------
  total                      3/94   59/179  74/210  84/133  31/50   16/15
  • 49 errors for BEIJING-LCG2_DATADISK [1]
  • 29 errors for GRIF-LAL_DATADISK [2]
  • 11 errors for GRIF-LAL_PRODDISK [3]
  • 56 errors for GRIF-SACLAY_DATADISK [4]
  • 21 errors for IN2P3-CPPM_DATADISK [5]
  • 15 errors for IN2P3-CPPM_PRODDISK [6]
  • 18 errors for IN2P3-LAPP_DATADISK [7]
  • 14 errors for IN2P3-LPC_DATADISK [8]
  • 8 errors for IN2P3-LPSC_DATADISK [9]
  • 15 errors for RO-02-NIPNE_DATADISK [10]
  • 6 errors for RO-07-NIPNE_DATADISK [11]
  • RO-07-NIPNE was in downtime[12], therefore zero success. But we still see this type of error in source preparation.
  • GRIF-LPNHE was in downtime [13] and FTS channel was active, therefore many destination errors. (zero success rate)
  • TOKYO was in downtime [14] and FTS channel was inactive, therefore no transfers.
  • Total 242 errors of this type between 12h00 - 16h00, mostly between 14h00-15h00.
  • there seem to be some inconsistency in the dashboard information. Thus the above numbers may not be exact.


Throughput (MB/s) [15] 
LYON.throughput.14400.20080805.1600.png
Number of files [16] 
LYON.num file xs.14400.20080805.1600.png
Number of transfer errors [17] 
LYON.num file xs error.14400.20080805.1600.png

Functional Test 11 Jun 2008

FDR

CCRC08