error running fad stats

[rhalperin]

I am getting the following error running fade stats:

rt/dwarfeh.d:354: uncaught exception reached top of stack
This might happen if you're missing a top level catch in your fiber or signal handler
core.exception.OutOfMemoryError@core/lifetime.d(137): Memory allocation failed
----------------
src/env/__libc_start_main.c:10 <ERROR: Unable to retrieve function name> [0x0]
Aborted

Here is how I am running fade:

docker run -it -m 8g -v $PWD:/data --entrypoint sh blachlylab/fade
fade stats $BAM

My host machine is a Mac. I've also been getting the same error on a cloud workstation running ubuntu 18.04 with fade installed via conda. I've tried several different bams including one that was trivially sized (0.5G) and have checked that I have plenty of free memory available on both machines. I am able to run fade out on the same bams without any errors.

[charlesgregory]

This error is misleading from the D language, fade stats $BAM should be run as fade stats $BAM output.tsv. Though there are other issues I found when investigating this so I will push up a new version in which fade stats $BAM works by writing to stdout.

[charlesgregory]

Can you try v0.5.6?

[rhalperin]

Running it as fade stats $BAM output.tsv worked. Thanks!

If you get the docker image and/or the conda installation updated to 0.5.6, then I will give it a try.

[charlesgregory]

The new docker image was pushed up as well with the release, you may have to run docker pull blachlylab/fade to get the latest version. conda will automatically update but it take probably a day or two for it to get worked into bioconda. If you get the same error, then there could be an issue with the docker image update process I use.

[rhalperin]

I pulled the latest docker, and I do see fade v0.5.6. However, I am getting a different error now:

fade stats $BAM
[E::hts_open_format] Failed to open file "stats" : No such file or directory
Segmentation fault

[charlesgregory]

Can you try v0.5.7? I found out in fixing this issue I introduced another which should now be fixed.

[rhalperin]

Yes, it appears fixed in 0.5.7, thanks!