Difficulty with malloc error

[rahulcdeo]

Hi,

I'm running Miso on Mac OS X 10.8.3 and am having frequent errors of this sort:

Python(57341) malloc: *** error for object 0x104120d00: incorrect checksum for freed object - object was probably modified after being freed.

I am running the --compute-genes-from-file step and looking at alternative splicing events.

Thanks

[yarden]

Hi,

What MISO version are you using? Did you try it with misopy-0.4.9 (latest)? Also, what version of GCC? Are you running it through any pipeline that uses make -j?

We've had a few users report this type of error, all on Mac OS X, but we've never been able to reproduce it, so any information you might have will be helpful. In the meantime, I suggest using Linux if possible.

--Yarden

[rahulcdeo]

Hi,

This is the latest github clone. OS X 10.8.3 no longer supports gcc4.2 so I used gcc4.7.

I examined the logfiles and noticed that there were only a handful of events (~5/1500) that were being considered that led to the crash. These all take place right after multiple reads are eliminated by filtering (same strand, no pairs, etc.,) - I wonder if there is some division by 0 problem that's occurring once filtering takes place.

I'm finding far fewer (if any) problems when I enable the pre-filter option. I'll switch over to my Linux cluster once I get Numpy and Scipy installed

Rahul

[gaborcsardi]

Hi, can you share a (preferably) small GFF and SAM/BAM file that reproduces the error? Thanks.

On Tue, Apr 30, 2013 at 10:23 AM, rahulcdeo [email protected]:

Hi,

This is the latest github clone. OS X 10.8.3 no longer supports gcc4.2 so I used gcc4.7.

I examined the logfiles and noticed that there were only a handful of events (~5/1500) that were being considered that led to the crash. These all take place right after multiple reads are eliminated by filtering (same strand, no pairs, etc.,) - I wonder if there is some division by 0 problem that's occurring once filtering takes place.

I'm finding far fewer (if any) problems when I enable the pre-filter option. I'll switch over to my Linux cluster once I get Numpy and Scipy installed

Rahul

— Reply to this email directly or view it on GitHubhttps://github.com/yarden/MISO/issues/46#issuecomment-17230021 .

[rahulcdeo]

Yes - I'll do so a little later today (am tied up in meetings). I should also have some of the log files which captured the crash.

Best, Rahul

On Apr 30, 2013, at 7:24 AM, Gabor Csardi [email protected] wrote:

Hi, can you share a (preferably) small GFF and SAM/BAM file that reproduces the error? Thanks.

On Tue, Apr 30, 2013 at 10:23 AM, rahulcdeo [email protected]:

Hi,

This is the latest github clone. OS X 10.8.3 no longer supports gcc4.2 so I used gcc4.7.

I examined the logfiles and noticed that there were only a handful of events (~5/1500) that were being considered that led to the crash. These all take place right after multiple reads are eliminated by filtering (same strand, no pairs, etc.,) - I wonder if there is some division by 0 problem that's occurring once filtering takes place.

I'm finding far fewer (if any) problems when I enable the pre-filter option. I'll switch over to my Linux cluster once I get Numpy and Scipy installed

Rahul

— Reply to this email directly or view it on GitHubhttps://github.com/yarden/MISO/issues/46#issuecomment-17230021 .

— Reply to this email directly or view it on GitHub.