Recorder Bug

[GiuseppeDelGobbo]

I am running many time history analyses in OpenSees using a loop to change the ground motions. I am finding that some of the recorders will be skipped and not record. This is always a random THA in my suite. For example I will run 50 THA, and for number 20 my interstorey drift recorder will not have created a file. I will run them again, and this time number 10 will not have a file created for the drift recorder. What could be happening to cause this? It is strange since the same files are being called each time.

When I rerun the loop but only with THA number = THA number with skipped recorder, it creates and records the file without problem.

[fmk]

it is interesting .. can you repeat with a sdof problem and post the script?

[GiuseppeDelGobbo]

Thanks for the reply. I will test this out and post the script/results.

[GiuseppeDelGobbo]

I have changed the model to a SDOF and no problems occur. The IDR file is created for each run. However, the IDR records do not record at t=0 when the record command is issues while all other recorders do.

I noticed that for the original/full model (in addition to some of the IDR files not being created) the IDR recorder starts at a random time (t=19 sec) rather than t=0sec for one analysis out of 50. Again, this only occurs for the IDR recorder with all other recorders starting at t=0 and creating a file for each analysis. Some IDR recorders record at t=0, while others start at t=dt

Could this be some sort of memory problem? If I rerun the script with the THA number specified, the IDR recorder starts at t=0 rather than t=19sec.