ku-awdc / runjags

An R package implementing interface utilities, model templates, parallel computing methods and additional distributions for MCMC models in JAGS
GNU General Public License v2.0
4 stars 2 forks source link

Issues retrieving temporary data to finish JAGS simulation #5

Open ecrjak opened 7 months ago

ecrjak commented 7 months ago

I am running a simulation via JAGS which continues to crash with the following message after running on an HPC system for several days. As advised by the error message, I am reporting this issue here. I have checked that there are no disk quota constraints as suggested by the error message. Thanks a lot in advance!

Adaptation successful
. NOTE: Stopping adaptation

. . . . . . . . . . . . Updating 2000
-------------------------------------------------| 2000
************************************************** 100%
. . . . . . Failed to open file sim.1/samplers.csv
. Updating 0
. Deleting model
NOTE: The JAGS output file(s) appear(s) to be very large - they may
take some time to read.  Have you accidentally included a large vector
in "monitor", or are you trying to run too many iterations without
specifying "thin"?If the read-in process fails (or is aborted), use ?results.jags and the
read.monitor argument to retrieve the simulationError in runjags.readin(directory = startinfo$directory, silent.jags = silent.jags,  : 
  Timed out waiting for the CODA files to be completed. You can wait for the files to be written, and then run results.jags("/tmp/RtmpsmrR6W/runjagsfiles1bb519123b944f")
The file size and modification times at 2024-02-11 23:52:09.883425 were: sim.1/CODAchain1.txt : 3860115456 : 2024-02-11 23:51:35.551446, sim.1/CODAchain2.txt : 1461170176 : 2024-02-11 23:51:35.551446, sim.1/CODAchain3.txt : 1461170176 : 2024-02-11 23:51:35.551446, sim.1/CODAchain4.txt : 1461170176 : 2024-02-11 23:51:35.551446.  Please file a bug report (including this message) to the runjags package author.
Calls: fx ... eforensics_main_par2 -> <Anonymous> -> runjags.readin
In addition: There were 50 or more warnings (use warnings() to see the first 50)
Note: Either one or more simulation(s) failed, or there was an error in
processing the results.  You may be able to retrieve any successful
simulations using:
results.jags("/tmp/RtmpsmrR6W/runjagsfiles1bb519123b944f",
recover.chains=TRUE)
See the help file for that function for possible options.
To remove failed simulation folders use cleanup.jags() - this will be
run automatically when the runjags package is unloaded
Error in gzfile(file, "wb") : cannot open the connection
Calls: fx ... eforensics_main_par2 -> <Anonymous> -> dumpmodelinfo -> save -> gzfile
In addition: Warning message:
In gzfile(file, "wb") :
  cannot open compressed file '/tmp/RtmpsmrR6W/runjagsfiles1bb519123b944f/jagsinfo.Rsave', probable reason 'Disk quota exceeded'
Execution halted
ecrjak commented 6 months ago

A quick follow up: the samplers.csv as well as the CODAchain.txt files do not exist in the sim.1 folder but in the one above. Is this some kind of bug or do I need to adjust some optional settings?