ORCHIDEE/OR4L OR4L - Historial de revisiones

Lluis.fita: Retoques

2017-11-07T18:37:59Z

Retoques

← Revisión anterior		Revisión del 18:37 7 nov 2017
Línea 140:		Línea 140:
	* '''NOTE:''' this will only work if one set-up the <code>.ssh</code> public/private keys in each involved USER/HOST.		* '''NOTE:''' this will only work if one set-up the <code>.ssh</code> public/private keys in each involved USER/HOST.
	* '''NOTE 2:''' All the forcings, compiled code, ... are already at <code>hydra</code> at the common space called <code>share</code>		* '''NOTE 2:''' All the forcings, compiled code, ... are already at <code>hydra</code> at the common space called <code>share</code>
	* '''NOTE 3:''' From the computing nodes, one can not access to the <code>/share</code> folder and to any of the CIMA's storage machines: skogul, freyja, ... For that reason, one need to use these system of <code>[USER]@[HOST]</code> accounts. <code>*.pbs</code> scripts uses a series of wrappers of the standard functions: <code>cp, ln, ls, mv, ....</code> which manage them `from' and `to' different pairs of <code>[USER]@[HOST]</code>		* '''NOTE 3:''' From the computing nodes, one can not access to the <code>/share</code> folder and to any of the CIMA's storage machines: skogul, freyja, ... For that reason, one need to use these system of <code>[USER]@[HOST]</code> accounts. <code>*.pbs</code> scripts uses a series of wrappers of the standard functions: <code>cp, ln, ls, mv, ....</code> which manage them `from' and `to' different pairs of <code>[USER]@[HOST]</code>. '''NOTE:''' this will only work if public keys have been set-up (see more details in [[llaves_ssh]])
	<pre>		<pre>
	# Hosts		# Hosts

Lluis.fita en 18:31 7 nov 2017

2017-11-07T18:31:57Z

← Revisión anterior		Revisión del 18:31 7 nov 2017
Línea 285:		Línea 285:
	</pre>		</pre>

	':' separated list of headers of netCDF file names from ORCHIDEE's output which need to be kept		':' separated list of headers of netCDF file names from ORCHIDEE's output which need to be kept ('''NOTE:''' if one of these files does not exist, system will stop!)
	<pre>		<pre>
	# Headers of netCDF files need to be kept		# Headers of netCDF files need to be kept

Lluis.fita: Rellenando la página OR4L IV

2017-10-18T19:22:49Z

Rellenando la página OR4L IV

@@ Línea 68: / Línea 68: @@
 $ ls -lrS out_orchidee_*
 </pre>
-* <code>run_or.log</code>: These are the files which content the standard output of the model. Search for `segmentation faults', etc...
+* <code>run_or.log</code>: These are the files which content the standard output of the model. Search for `segmentation faults' in form of (it might differ):
 * on <code>[runHOME]/[ExpName]/[SimName]</code>, check the output of the PBS jobs. Which are called:
 ** <code>exp_oF-[SimName].o[nnnn]</code>: output of the <code>run_experiment.pbs</code>

Lluis.fita: Rellenando la página OR4L III

2017-10-18T19:18:19Z

Rellenando la página OR4L III

← Revisión anterior		Revisión del 19:18 18 oct 2017
Línea 70:		Línea 70:
	* <code>run_or.log</code>: These are the files which content the standard output of the model. Search for `segmentation faults', etc...		* <code>run_or.log</code>: These are the files which content the standard output of the model. Search for `segmentation faults', etc...
	* on <code>[runHOME]/[ExpName]/[SimName]</code>, check the output of the PBS jobs. Which are called:		* on <code>[runHOME]/[ExpName]/[SimName]</code>, check the output of the PBS jobs. Which are called:
	* <code>exp_oF-[SimName].o[nnnn]</code>: output of the <code>run_experiment.pbs</code>		** <code>exp_oF-[SimName].o[nnnn]</code>: output of the <code>run_experiment.pbs</code>
	* <code>or_oF-[SimName].o[nnnn]</code>: output of the <code>run_OR.pbs</code>		** <code>or_oF-[SimName].o[nnnn]</code>: output of the <code>run_OR.pbs</code>
			* Check <code>[runHOME]/[ExpName]/[SimName]/run/used_run.def</code> which holds all the parameters (even the default ones) used in the simulation

	== EXPERIMENTSparameters.txt ==		== EXPERIMENTSparameters.txt ==

Lluis.fita: Añadiendo qué buscar cuando hay errores II

2017-10-18T19:14:57Z

Añadiendo qué buscar cuando hay errores II

← Revisión anterior		Revisión del 19:14 18 oct 2017
Línea 69:		Línea 69:
	</pre>		</pre>
	* <code>run_or.log</code>: These are the files which content the standard output of the model. Search for `segmentation faults', etc...		* <code>run_or.log</code>: These are the files which content the standard output of the model. Search for `segmentation faults', etc...
			* on <code>[runHOME]/[ExpName]/[SimName]</code>, check the output of the PBS jobs. Which are called:
			* <code>exp_oF-[SimName].o[nnnn]</code>: output of the <code>run_experiment.pbs</code>
			* <code>or_oF-[SimName].o[nnnn]</code>: output of the <code>run_OR.pbs</code>

	== EXPERIMENTSparameters.txt ==		== EXPERIMENTSparameters.txt ==

Lluis.fita: Añadiendo qué buscar cuando hay errores

2017-10-18T19:07:30Z

Añadiendo qué buscar cuando hay errores

← Revisión anterior		Revisión del 19:07 18 oct 2017
Línea 59:		Línea 59:
	* <code>[runHOME]/[ExpName]/[SimName]</code>: Will content the copies of the templates <code>run.def</code>, <code>*.xml</code> and a file <code>chunk_attemps.inf</code> which counts how many times a '''chunk''' has been attempted to be run (if it reached 4 times, the <code>OPR4L</code> is stopped)		* <code>[runHOME]/[ExpName]/[SimName]</code>: Will content the copies of the templates <code>run.def</code>, <code>*.xml</code> and a file <code>chunk_attemps.inf</code> which counts how many times a '''chunk''' has been attempted to be run (if it reached 4 times, the <code>OPR4L</code> is stopped)
	* <code>[runHOME]/[ExpName]/[SimName]/run</code>: actual folder where the computing nodes run the model. In a folder called <code>orout</code> there is a folder for each '''chunk''' with the standard output of the model		* <code>[runHOME]/[ExpName]/[SimName]/run</code>: actual folder where the computing nodes run the model. In a folder called <code>orout</code> there is a folder for each '''chunk''' with the standard output of the model
			* <code>[runHOME]/[ExpName]/[SimName]/run/orout/[YYYYi][MMi][DDi][HHi][MIi][SSi]-[YYYYf][MMf][DDf][HHf][MIf][SSf]</code>: folder with the standard output and all the required files to run a given '''chunk'''. The content of all this folder is compressed and kept in <code>[storageHOME]/[ExpName]/[SimName]/config_[YYYYi][MMi][DDi][HHi][MIi][SSi]-[YYYYf][MMf][DDf][HHf][MIf][SSf].tar.gz</code>
	* <code>[storageHOME]/[ExpName]/[SimName]</code> (in [storageHOST]): output of the already ran '''chunks''' as <code>[YYYYi][MMi][DDi][HHi][MIi][SSi]-[YYYYf][MMf][DDf][HHf][MIf][SSf]</code> for a chunk from <code>[YYYYi]/[MMi]/[DDi] [HHi]:[MIi]:[SSi]</code> to <code>[YYYYf]/[MMf]/[DDf] [HHf]:[MIf]:[SSf]</code>		* <code>[storageHOME]/[ExpName]/[SimName]</code> (in [storageHOST]): output of the already ran '''chunks''' as <code>[YYYYi][MMi][DDi][HHi][MIi][SSi]-[YYYYf][MMf][DDf][HHf][MIf][SSf]</code> for a chunk from <code>[YYYYi]/[MMi]/[DDi] [HHi]:[MIi]:[SSi]</code> to <code>[YYYYf]/[MMf]/[DDf] [HHf]:[MIf]:[SSf]</code>
	* If there has been any problem, check the last chunk and ~~its standard output (~~<code>out_orchidee_[nnnn]</code> files) which ~~is kept in~~ a ~~compressed file~~ <code>~~[storageHOME]/[ExpName]/[SimName]/config_[YYYYi][MMi][DDi][HHi][MIi][SSi]-[YYYYf][MMf][DDf][HHf][MIf][SSf].tar~~.gz</code> ~~to try to understand what happens and where~~ the ~~problem comes from~~
			=== When something went wrong ===
			If there has been any problem check the last chunk (in orout/[PERIODchunk]) to try to understand what happens and where the problem comes from:
			* <code>out_orchidee_[nnnn]</code>: These are the files which content the standard output while running the model. One file for each process. If the problem was something related to model execution and it has been prepared for the error, a correct message must appear. (look first for the largest files...
			<pre>
			$ ls -lrS out_orchidee_*
			</pre>
			* <code>run_or.log</code>: These are the files which content the standard output of the model. Search for `segmentation faults', etc...

	== EXPERIMENTSparameters.txt ==		== EXPERIMENTSparameters.txt ==

Lluis.fita en 19:49 11 oct 2017

2017-10-11T19:49:35Z

@@ Línea 38: / Línea 38: @@
 $ qsub run_experiment.pbs
 </pre>
 In case of crash of the simulation, after fixing the issue, go to <code>[runHOME]/[ExpName]/[SimName]</code> and re-launch the experiment (after the first run the <code>scratch</code> is switched automatically to `false')
 <pre>

Lluis.fita: Rellenando la página OR4L re-launch

2017-10-11T19:44:05Z

Rellenando la página OR4L re-launch

← Revisión anterior		Revisión del 19:44 11 oct 2017
Línea 35:		Línea 35:
	</pre>		</pre>
	# Launch the simulation of the experiment		# Launch the simulation of the experiment
			<pre>
			$ qsub run_experiment.pbs
			</pre>
			In case of crash of the simulation, after fixing the issue, go to <code>[runHOME]/[ExpName]/[SimName]</code> and re-launch the experiment (after the first run the <code>scratch</code> is switched automatically to `false')
	<pre>		<pre>
	$ qsub run_experiment.pbs		$ qsub run_experiment.pbs

Lluis.fita: Añadidiendo chequeo de parado de simulaciones

2017-10-11T19:36:23Z

Añadidiendo chequeo de parado de simulaciones

← Revisión anterior		Revisión del 19:36 11 oct 2017
Línea 44:		Línea 44:
	* <code>[runHOME]/[ExpName]/[SimName]/run</code>: actual folder where the computing nodes run the model. In a folder called <code>orout</code> there is a folder for each '''chunk''' with the standard output of the model		* <code>[runHOME]/[ExpName]/[SimName]/run</code>: actual folder where the computing nodes run the model. In a folder called <code>orout</code> there is a folder for each '''chunk''' with the standard output of the model
	* <code>[storageHOME]/[ExpName]/[SimName]</code> (in [storageHOST]): output of the already ran '''chunks''' as <code>[YYYYi][MMi][DDi][HHi][MIi][SSi]-[YYYYf][MMf][DDf][HHf][MIf][SSf]</code> for a chunk from <code>[YYYYi]/[MMi]/[DDi] [HHi]:[MIi]:[SSi]</code> to <code>[YYYYf]/[MMf]/[DDf] [HHf]:[MIf]:[SSf]</code>		* <code>[storageHOME]/[ExpName]/[SimName]</code> (in [storageHOST]): output of the already ran '''chunks''' as <code>[YYYYi][MMi][DDi][HHi][MIi][SSi]-[YYYYf][MMf][DDf][HHf][MIf][SSf]</code> for a chunk from <code>[YYYYi]/[MMi]/[DDi] [HHi]:[MIi]:[SSi]</code> to <code>[YYYYf]/[MMf]/[DDf] [HHf]:[MIf]:[SSf]</code>
			* If there has been any problem, check the last chunk and its standard output (<code>out_orchidee_[nnnn]</code> files) which is kept in a compressed file <code>[storageHOME]/[ExpName]/[SimName]/config_[YYYYi][MMi][DDi][HHi][MIi][SSi]-[YYYYf][MMf][DDf][HHf][MIf][SSf].tar.gz</code> to try to understand what happens and where the problem comes from

	== EXPERIMENTSparameters.txt ==		== EXPERIMENTSparameters.txt ==

Lluis.fita: Rellenando la página OR4L en cómo seguir la simulación

2017-10-11T19:25:52Z

Rellenando la página OR4L en cómo seguir la simulación

← Revisión anterior		Revisión del 19:25 11 oct 2017
Línea 39:		Línea 39:
	</pre>		</pre>

			== Checking the experiment ==
			Once the experiment runs, one needs to look on (following name of the variables from <code>EXPERIMENTparameters.txt</code>
			* <code>[runHOME]/[ExpName]/[SimName]</code>: Will content the copies of the templates <code>run.def</code>, <code>*.xml</code> and a file <code>chunk_attemps.inf</code> which counts how many times a '''chunk''' has been attempted to be run (if it reached 4 times, the <code>OPR4L</code> is stopped)
			* <code>[runHOME]/[ExpName]/[SimName]/run</code>: actual folder where the computing nodes run the model. In a folder called <code>orout</code> there is a folder for each '''chunk''' with the standard output of the model
			* <code>[storageHOME]/[ExpName]/[SimName]</code> (in [storageHOST]): output of the already ran '''chunks''' as <code>[YYYYi][MMi][DDi][HHi][MIi][SSi]-[YYYYf][MMf][DDf][HHf][MIf][SSf]</code> for a chunk from <code>[YYYYi]/[MMi]/[DDi] [HHi]:[MIi]:[SSi]</code> to <code>[YYYYf]/[MMf]/[DDf] [HHf]:[MIf]:[SSf]</code>

	== EXPERIMENTSparameters.txt ==		== EXPERIMENTSparameters.txt ==